[sdiy] TX81Z, YMxx chips, and aliasing

[Pomik]

Hi, 
my first comment here...
I have had big stress with my DX7 mk.I. digital sig noise,

since I realised it on Headphones 

after some months from a spawnshop buy.

I Googled a lot, found schematics,

also looked over forums around,

and if anybody ask for antialiasing,?

Isnt this question about more or less filtering

of output from DAC ? 

Cause of ol original DX7 noise for example

is that thing of switching MOS 4 bit R2R add-on 

for 12 bit ADC, and S-H for 4(6 ?) chanell of voise,




Thinking bout upgrading it for a cleaner sound of DX7 Mk II/ D FD

found that these have S-H swithed to stereo out from one single PCM57 ADC.




but signals and OPGenerator  out wiring are quite different ?(










Antialiasing algorithms in FM ?

Gues it as so DSP hungry and difficult,

that original FM synth buiders didnt even think bout that.

-- 
s pozdravem 
Petr M 
(+420)608228257 ...<<<<<<< *<:]<> Pom.ixsss...

---------- Původní e-mail ----------
Od: KD KD <pic24hj at gmail.com>
Komu: synth-diy at synth-diy.org
Datum: 8. 9. 2018 0:56:55
Předmět: Re: [sdiy] TX81Z, YMxx chips, and aliasing 
"Yes, i think all yamaha PM synths have keyboard scaling to change
modulation index.
e.g page 144 SY99 manual 4 breakpoints describes how the level of a
modulator or operator over keyboard are freely scaled. All waveforms
are mathematical transforms
of sines.

2018-09-05 14:14 GMT+02:00, rburnett at richieburnett.co.uk
<rburnett at richieburnett.co.uk>:
> As Michael said, the FM spectrum is theoretically infinite even with a
> sinusoidal carrier and modulator.
>
> However, I believe that synths like the DX-7 reduced audible aliasing by
> employing mappings across the keyboard that decreased the modulation
> index of patches progressively for higher pitched notes. Reducing the
> modulation index in Frequency Modulation puts less energy into the
> higher-order sidebands and more energy into the close-in sidebands from
> the FM process, so it results in a signal that has less potential to
> alias badly.
>
> -Richie,
>
> On 2018-09-05 12:56, Declare Update wrote:
>> Ah! very good point. thanks.
>>
>> Chris
>>
>>> On Sep 5, 2018, at 4:14 PM, Michael Zacherl <sdiy-mz01 at blauwurf.info>
>>> wrote:
>>>> On 3 Sep 2018, at 09:43, Declare Update <declareupdate at gmail.com>
>>>> wrote:
>>>> Do the yamaha FM chips, such as the YM2414 in the TX81Z, do anything
>>>> toward anti-aliasing? more specifically, does anyone know if they use
>>>> band limited tables for their extra wave forms besides sine, and is
>>>> there any attempt to band limit the signal after FM is applied?
>>>
>>> On the bottomline bandlimiting the source wave forms is moot, since
>>> it’s the FM’s sidebands which cause the aliasing (and are responsible
>>> for the rich spectrum).
>>> So even when using just sine waves you may get side bands which get
>>> above Nyquist and/or below zero.
>>> m.

>

_______________________________________________
Synth-diy mailing list
Synth-diy at synth-diy.org
http://synth-diy.org/mailman/listinfo/synth-diy
"
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://synth-diy.org/pipermail/synth-diy/attachments/20180912/f28ed257/attachment.htm>