dillon@CORY.BERKELEY.EDU (Matt Dillon) (05/16/88)
>The Amiga Hardware Reference Manual says that 29 KHz is the maximum >playback rate for audio samples. Is this for sure true and are there >any sorts of software-achievable hacks to exceed it? >-- >"Now here's something you're really going to like!" -- Rocket J. Squirrel >...!{bellcore!tness1,uunet!nuchat}!sugar!karl, Unix BBS (713) 438-5018 From what I understand, 31Khz is possible, but the hardware was really designed for 28.867Khz to 'save buffers'. The hardware manual states that the audio DMA cannot keep up with the rate at higher frequencies. -Matt
grr@cbmvax.UUCP (George Robbins) (05/16/88)
In article <1991@sugar.UUCP> karl@sugar.UUCP (Karl Lehenbauer) writes: > The Amiga Hardware Reference Manual says that 29 KHz is the maximum > playback rate for audio samples. Is this for sure true and are there > any sorts of software-achievable hacks to exceed it? Well, you can always load the audio data registers programmatically, however if you want to do this with any degree of frequency and accuracy you will have to take over the machine and perhaps even disable video/DMA. Would make for some interesting demos, however it's not clear that the Amiga audio circuitry is of sufficient quality to really benefit from faster sampling rate - definitly not with the normal anti-aliasing filter enabled. -- George Robbins - now working for, uucp: {uunet|ihnp4|rutgers}!cbmvax!grr but no way officially representing arpa: cbmvax!grr@uunet.uu.net Commodore, Engineering Department fone: 215-431-9255 (only by moonlite)
sterling@cbmvax.UUCP (Rick Sterling QA) (05/16/88)
In article <8805151814.AA08523@cory.Berkeley.EDU> dillon@CORY.BERKELEY.EDU (Matt Dillon) writes: > > >The Amiga Hardware Reference Manual says that 29 KHz is the maximum > >playback rate for audio samples. Is this for sure true and are there > >any sorts of software-achievable hacks to exceed it? > >-- > >"Now here's something you're really going to like!" -- Rocket J. Squirrel > >...!{bellcore!tness1,uunet!nuchat}!sugar!karl, Unix BBS (713) 438-5018 > > From what I understand, 31Khz is possible, but the hardware was > really designed for 28.867Khz to 'save buffers'. The hardware manual > states that the audio DMA cannot keep up with the rate at higher frequencies. > > -Matt This is true for DMA controlled audio playback only. If you are willing to take direct control of the D/A's with the CPU and some machine code you can get output up in the megahertz range. ( You'd have to disable the built in anti-aliasing low-pass filters though. ) ============================================================================= Rick Sterling COMMODORE AMIGA TEST ENGINEERING // /_ |\/||/_ /_ UUCP ...{allegra,ihnp4,rutgers}!cbmvax!sterling \X/ / \| ||\// \ PHONE 215-431-9275 ============================================================================= Everybody likes hard work ... especially when THEY'RE paying for it. =============================================================================
elg@killer.UUCP (Eric Green) (05/17/88)
in article <1991@sugar.UUCP>, karl@sugar.UUCP (Karl Lehenbauer) says: > The Amiga Hardware Reference Manual says that 29 KHz is the maximum > playback rate for audio samples. Is this for sure true and are there > any sorts of software-achievable hacks to exceed it? 1) Most good-quality stereos give up the ghost somewhere around 20khz. I'd be surprised if the monitor on your speaker could make it past 10khz. 2) Most adult's ears give up around 17-18khz (for someone with VERY good hearing). If you cannot hear the 15khz whistle that your monitor is giving off this very second..... Sorry, but I think this is a non-problem. -- Eric Lee Green {cuae2,ihnp4}!killer!elg Snail Mail P.O. Box 92191 Lafayette, LA 70509 "Is a dream a lie that don't come true, or is it something worse?"
mat@emcard.UUCP (Mat Waites) (05/17/88)
In article <4113@killer.UUCP> elg@killer.UUCP (Eric Green) writes: >in article <1991@sugar.UUCP>, karl@sugar.UUCP (Karl Lehenbauer) says: >> The Amiga Hardware Reference Manual says that 29 KHz is the maximum >> playback rate for audio samples. Is this for sure true and are there >> any sorts of software-achievable hacks to exceed it? > >1) Most good-quality stereos give up the ghost somewhere around 20khz. [...] >2) Most adult's ears give up around 17-18khz (for someone with VERY good [...] > >Sorry, but I think this is a non-problem. > >-- > Eric Lee Green {cuae2,ihnp4}!killer!elg BUT, you have to sample at a significantly greater rate than the frequency you are trying to reproduce. Don't cd players sample at 64khz or something like that just to get good reproduction at 20khz? I think the rule of thumb is that the minimum sampling rate is twice the frequency you are trying to reproduce; 29khz sampling will only return a clean sound for sounds below 14.5khz by that rule. 14.5 is still pretty good unless you are trying for hi-fi. The higher frequencies are not necessarily needed for actual tones in the music, but for maintaining the ambience or spatial quality of the music. And even old folks who can't "hear" a pure 18khz tone can tell the difference between music that has the highs filtered out, and music that doesn't. mat -- W Mat Waites | PHONE: (404) 727-7197 Emory Univ Cardiac Data Bank | UUCP: ...!gatech!emcard!mat Atlanta, GA 30322 |
haitex@pnet01.cts.com (Wade Bickel) (05/18/88)
First of all, ANYONE WHO HAS NOT THOUROUGLY INVESTIGATED THE
NATURE OF SOMEONE ELSE'S REQUEST FOR INFORMATION HAS ABSOLUTLY NO
BUISNESS CLAIMING THAT SUCH A PERSONS' REQUEST IS UNFOUNDED. THIS
WAS MADE PAINFULLY CLEAR IN THIS THREAD WHEN ONE PERSON WHO DID NOT
EVEN UNDERSTAND THAT IN ORDER TO SAMPLE A WAVEFORM OF A GIVEN
FREQUENCY THE SAMPLES *MUST* BE TAKEN AT TWICE THAT FREQENCY.
FAILURE TO DO SO RESULTS IN GROSS ALIASING WHICH IS QUITE NOTICEABLE
AT SAMPLING RATES BELOW ABOUT 40KHZ.
THIS IS A PLACE TO OFFER POSITIVE SUGGESTIONS OR CRITICISMS.
===== BREIF FLAME OFF =====
Also, I happen to have started the laborious task of reading my
DevCon notes last night, and just happend to run accross the
following:
(found in section 1, actuall page 16, titledt a the bottom
"Outline of Features in V1.3, V1.4 and the Enhanced Custom Chips"
"Page 3"
Agnus
1).....
Audio Channel period: The minimum period is 124 color clocks,
which gives a maximum sample frequency of 28.86 khz. However,
if the horizontal frequency is different, the sampling freq.
can be higher. The minimum period is (Htotal+21)/2 color
clocks.
)
Of course this may not apply to current rev.s of Agnus, but it
might??? If it is possible to alter the horizontal freq. then this
might be a way to achieve a higher sampling rate, at the expense of
the display (???).
Hope this is usefull to someone.
Thanks,
Wade.
UUCP: {cbosgd, hplabs!hp-sdd, sdcsvax, nosc}!crash!pnet01!haitex
ARPA: crash!pnet01!haitex@nosc.mil
INET: haitex@pnet01.CTS.COMcmcmanis@pepper.UUCP (05/18/88)
In article <8805171810.AA07554@cory.Berkeley.EDU> (Matt Dillon) writes: > I think everybody is missing a major point here... If the limit >to the audio system is 29KHz/channel, it means that the highest frequency >you can get out of the thing is about 14KHz... and that would HAVE to be a >square wave! > -Matt Matt, this is what the low-pass filter is for. You are correct that the maximum frequency that this setup can produce is 14Khz. However on the output is a low pass filter that filters out any frequencies above 7Khz or so. Ideally this would have a vertical cutoff at 14Khz so that when you send a wave of 0xff 0x00 to the output the filter removes all of the 'artifacts' above 14Khz and you get a pure 14Khz sinewave out. Sudden transitions in the output are heard as 'quantization' noise and are generally much higher frequencies than the waveform you are trying to produce. That's why a filter is essential in any digital sound synthesis system. The end result is that a two byte sample to play a sine wave at 10Khz has just as much fidelity as a 256 byte sample at 78Hz would. --Chuck McManis uucp: {anywhere}!sun!cmcmanis BIX: cmcmanis ARPAnet: cmcmanis@sun.com These opinions are my own and no one elses, but you knew that didn't you.
dillon@CORY.BERKELEY.EDU (Matt Dillon) (05/19/88)
:In article <8805171810.AA07554@cory.Berkeley.EDU> (Matt Dillon) writes:
:> I think everybody is missing a major point here... If the limit
:>to the audio system is 29KHz/channel, it means that the highest frequency
:>you can get out of the thing is about 14KHz... and that would HAVE to be a
:>square wave!
:> -Matt
:
:Matt, this is what the low-pass filter is for. You are correct that the
:maximum frequency that this setup can produce is 14Khz. However on the
:output is a low pass filter that filters out any frequencies above 7Khz
:or so. Ideally this would have a vertical cutoff at 14Khz so that when
:you send a wave of 0xff 0x00 to the output the filter removes all of
:the 'artifacts' above 14Khz and you get a pure 14Khz sinewave out.
I stand corrected, though my meaning was more towards 'You can't
have fancy waveforms at 14Khz because you have no room to play with since
you must do it in essentially two samples'.
-Mattvkr@osupyr.mast.ohio-state.edu (Vidhyanath K. Rao) (05/21/88)
Ignore the engineering problems. If the repproducing equipment has a +/-1dB response from 20Hz to 15KHz, but only +/-10dB at 20KHz, how many can tell the difference from some thing that is +/-3dB upto 20KHz? This is idle curiosity when applied to analog equipment (read speakers); but has such a test been ever done?