| |
Dr. 8 Bit Account closed
Registered: Sep 2006 Posts: 25 |
reduce the audio-noise
How can I reduce the audio-noise from c64?? some ideas?? a capcitor...mmm.. |
|
| |
Rough Account closed
Registered: Feb 2002 Posts: 1829 |
turn down the volume...
yes, I know, not funny. |
| |
Scout
Registered: Dec 2002 Posts: 1568 |
Quote: How can I reduce the audio-noise from c64?? some ideas?? a capcitor...mmm..
Questions, questions, questions...
Every day a new question by Dr. 8Bit.
Try to collect them all! \o/ |
| |
Mace
Registered: May 2002 Posts: 1799 |
Dr. 8 bit, you should become a member of the http://www.prophet64-forum.com.
There you will find most of the answers to your questions.
But a quick reply: build a controlled feedback loop through the SID input of your C64. |
| |
Raf
Registered: Nov 2003 Posts: 343 |
what mace said :)
search for mr AlphA 's modded C64-II he uses in his studio for prophet 64 exclusively - he provides schematics . charts and audio samples of solutions he proposed.
www.vulture.c64.org |
| |
MagerValp
Registered: Dec 2001 Posts: 1056 |
What if you just ground the SID input?
|
| |
Bamu® Account closed
Registered: May 2005 Posts: 1332 |
I usually remove the noise with some pc software. |
| |
uneksija
Registered: Jan 2002 Posts: 44 |
Quote: What if you just ground the SID input?
Hmm yes I think this was mentioned on Alpha's site as an effective way to remove a lot of the noise. However, do it at your own risk :P
http://www.bigmech.com/misc/c64mods/noisereduction.html |
| |
Grue
Registered: Dec 2001 Posts: 146 |
http://www.sid.fi/~grue/temp/hifi_sixtyfour2.txt
This method is pretty effective atleast on 6581 chips, it insulates 6581 from the data and address busses when CS (Chip select) line for sid is not active.
Combined with filtered and separate +12v and +5v voltages and own separate grounding it improves output quality alot. Also grounding potx, poty and ext-in lines with small (1nF) capasitor helps littlebit on the way. Also building separate audio out buffer (pre amp) and making own audio out connector helps on the way too.
|
| |
null Account closed
Registered: Jun 2006 Posts: 645 |
I say:
Noise == <3
------------------------------------
Knoeki/DigitalSoundsSystem |
| |
Mace
Registered: May 2002 Posts: 1799 |
Some people remove noise.
Others make a distortion plugin.
:-) |
| |
dalezy
Registered: Apr 2002 Posts: 475 |
noise == <3, so true.
yet, hardware-buzzing coming from the vic with only one goal in mind: making the recording sound fucked up != <3.
i'd prefer to do it the software way, yet i'm not quite the best among the mastering-bunch. is there any surefire way to get rid of *any* bastard-sounds without hearable loss of the intended sound? |
| |
Hate Bush
Registered: Jul 2002 Posts: 453 |
@pan engmann:
you mean, the software way? no. if signal and noise share the same frequency, there is little you can do about it. (otherwise, it should be easy enough.) noise profile based method fails miserably in this case, you'll be better off with manual eq'ing the noise out, using a few thin notch filters.
the best way to get around the noise being 'under' the signal is to fatten up the arrangement, in particular: introducing more tracks operating in required frequency area. for example, a 8-14kHz hiss on vocals can be easily 'patched' by glossy, sharp hihat.
of course, there are professional mastering tools that make the process a breeze, more or less. however i doubt if you can afford them.
care to correct me, someone? i'd be more than glad. |
| |
Bamu® Account closed
Registered: May 2005 Posts: 1332 |
Software way: I usually use Goldwave for this. The only thing that is required is a background-noise sample. The result is usually extremely good (good enough for audio-cds). |
| |
Bamu® Account closed
Registered: May 2005 Posts: 1332 |
Here two examples. It removes the most noise, but not everything.
Anyway, it's a good solution instead of hacking the hardware. :)
http://www.rogepost.com/n/9999495240
|
| |
gregg Account closed
Registered: Apr 2005 Posts: 56 |
I think it's common knowledge, but just in case: disable VIC while playing, it'll get rid of much noise. The MMC64 browser is capable of that (just press space).
Noise removal postprocessing (I did it with Audacity) always introduced some kind of artefacts for me. |
| |
chatGPZ
Registered: Dec 2001 Posts: 11119 |
especially when it comes to the sid, using postprocessing isnt very effective. better eliminate the noise by other means and dont record it in the first place. |
| |
Hate Bush
Registered: Jul 2002 Posts: 453 |
is it possible to disable the VIC without MMC64? if so, how to do it? |
| |
Oswald
Registered: Apr 2002 Posts: 5018 |
poke53265,0 |
| |
JackAsser
Registered: Jun 2002 Posts: |
Quote: poke53265,0
Dunno if you're were trying to make a joke Oswald.. :D Anyways, that poke doesn't disable the horizontal syncs, hence the high pitched noise will remain. On my C64 where this noise was rather extreme I insulated the h-sync signal and copper taped the whole VIC-cage. This helped very much. IIRC my boss added some sort of filter as well + did something with the line-in on the SID-chip. I'll ask him when he comes back. |
| |
WVL
Registered: Mar 2002 Posts: 886 |
Quote: is it possible to disable the VIC without MMC64? if so, how to do it?
lda #$0b
sta $d011
...
The 'noise' mainly comes from data flowing through the channels. But I'm wondering if this really helps? Isnt the design of the Vic such that it _must_ fetch a byte every cycle? I think the best would be to simply fill up a bank completely with 00's and let the vic display that, aswell as turning the screen off.
Then I'd let the 6502 do as little as possible by only doing lda #$a9 (or similar) when you don't have to jsr to the tune.. |
| |
Oswald
Registered: Apr 2002 Posts: 5018 |
jack, well if pressing space on the mmc64 equals turning the display enable bit 0, then my answer is correct I guess :) |
| |
aeeben
Registered: May 2002 Posts: 42 |
lda #$08
sta $d417
|
| |
JackAsser
Registered: Jun 2002 Posts: |
Quote: lda #$08
sta $d417
Ok... what noise are we talking about here? The high-pitched noise or some general white noise?
The high-pitched noise comes from the h-syncs for a fact, about 15625Hz. General white noise I dunno where that comes from.
Poking into $d011 with any value will NEVER change the hsync behavior (unless it's a C128).
@Oswald: Yeah, got your point now.. ;D |
| |
gregg Account closed
Registered: Apr 2005 Posts: 56 |
I dunno why, but setting DEN=0 gets rid of a lot of low-pitched noise... they hsync noise doesn't matter much for me. It's probably possible to cutoff at 15Khz without any real side-effects anyway. |
| |
Devia
Registered: Oct 2004 Posts: 401 |
Actually, disabling DEN and setting $d020 to 0, will reduce the h-sync noise considerably. Setting $d020 and $d021 to 0 and clearing $0400 will do just about the same. The point is that there should be no pixel changes to the display. Adding random noise to the screen increases the noise and it seems the "lighter" the colour of the screen-noise, the louder the audible noise.
It's been a while since I did experiments on this, but please do correct me if I'm remembering something wrong here ;-)
|
| |
chancer
Registered: Apr 2003 Posts: 343 |
depends in what sense you mean exactly? to record to audio etc. or just generally.
if its audio as mason said before, you could just run an EQ cut on the sound.. and that will eliminate the frequencies.. e.g. anything under say 30-40hz or the high end.. so it gates the sound.
another way of doing that is running it thru a graphic eq, and boosting / muting certain frequencies.
as for the chips, hhm just before its going to audio.. it's either capacitors or resistors, i forget which, although i'm sure its capacitors.. but put them on the cable OUT of the c64.. again that acts like a gate. that techique i used with car audio, so although the cross over will seperate the signal in frequencies, the mid range, had a capacitor and the rest went to the tweeters.
|
| |
MagerValp
Registered: Dec 2001 Posts: 1056 |
Again, grounding the SID input line, preferably via a 100 Ohm resistor or so, makes a the SID a lot less susceptible to picking up noise. A blank, black screen makes the VIC produce less noise. Combine as needed.
|
| |
blinkenlichten Account closed
Registered: Aug 2003 Posts: 4 |
Yes, bad grounding is most likely the reason for your high freq noise. This is from my experience with my record players, but the principle should apply to any audio equipment. |
| |
MagerValp
Registered: Dec 2001 Posts: 1056 |
On your average C64 the SID input is unconnected and floating, and basically acting as an antenna. Not good.
|
| |
chatGPZ
Registered: Dec 2001 Posts: 11119 |
depending where you live, you can stick a capacitor in there and listen to AM radio =) |
| |
uneksija
Registered: Jan 2002 Posts: 44 |
I've tried the "ground the input"-trick. I didn't use any resistor, I simply used a shorted plug (so basically, I wasn't even hacking the hardware). The results were amazing. Almost all of the noise disappeared. No need for any other tricks... |
| |
gregg Account closed
Registered: Apr 2005 Posts: 56 |
As long as you don't short-cut the internal capacitor that is in the audio-in line, it should be safe.
I simply soldered both pins in the DIN-plug together, they are right next to each other.
So, put audio-in to ground, lda #0 sta $d011 sta $d020 and no noise whatsoever anymore for me. :) |
| |
Hate Bush
Registered: Jul 2002 Posts: 453 |
uneksija, gregg: how to _exactly_ modify the plug? care to tell me in details? sadly, i'm an idiot when it comes to any modding. |
| |
dalezy
Registered: Apr 2002 Posts: 475 |
has anyone ever tried to plug a separator filter between the c64 and the recording target? i used such a thing before to filter out the noise in my amp that was caused by my screen, time to search for that cable and try it out =) |
| |
Devia
Registered: Oct 2004 Posts: 401 |
Randall: Just open up the DIN plug and short the two pins labeled 2 and 5
|
| |
Frantic
Registered: Mar 2003 Posts: 1627 |
I guess this is in accordance (somehow?) with what me and Twoflower found while doing some sid tests once. We found that reading voice 3 when playing combined waveform 5 had slightly different results on different 8580-machines. Some produced more high values (above $80) than others. We didn't do it very carefully (i.e. "sampling" each cycle with a REU) but bascially just did some non timed polling, but the results were consistent at least. Machines known as "skärmtjutare" in swedish (made in Hong Kong or something?) = machines that produce a lot of audio noise (a "hissing noise") when the screen is bright, also produced more "high values" in voice three. I guess it was basically some kind of noise, rather than the waveform actually being different in these chips.
We dealed with this when investigating SID chip detection, but since then I learned better ways of detecting the sid anyway, and never cared much about checking it further.
However, I guess someone might have some ideas about this. I suck at hardware, so I don't really have any theories.
Perhaps this "feature" can be used to detect wheter a digital android is using the machine (i.e. Iopop and others'), rather than a analogue human (which wouldn't cause any interrrferrrence with the digital equipment)? ;) |
| |
uneksija
Registered: Jan 2002 Posts: 44 |
...Or do it like me: I've got an adapter cable that has the DIN plug on the other end and four female RCA's on the other end. One of the RCA's is the audio output and one is the audio input. I made a special plug for grounding the audio input. The plug is simply a male RCA plug, which I had shorted by soldering a little wire (a piece of a component leg actually) between the signal and ground of the plug. I'd send a picture which would really give you the idea, but I don't have my plug here at the moment. This special plug I insert to the audio input (while C64 is turned OFF!) and ta dah! Noise is gone! I've heard this might also boost digis on an 8580 (don't know if it's true), but I haven't tried it, because my old C64C has probelms with its TV-signal (Will fix this soon). |
| |
ready.
Registered: Feb 2003 Posts: 441 |
No, this will not boost any digis on 8580 model. In fact the right way to implement this is to ground the actual SID pin via a resistor to ground. Remeber that between AUDIO IN pin on SID and AUDIO IN on connector on the rear side of the C64c, there is a capacitor to ground.
You have to put the resistor between sid pin and capacitor, otherwise you will not get $d814-digi replay.
Anyways this modification has to be done on the C64 board (I actually soldered on the 8580 pin) and it works great. Except you loose the feature that 8580 does not produce click when changing the volume $d814.
|
| |
Sledge
Registered: Sep 2003 Posts: 102 |
http://www.bigmech.com/misc/c64mods/noisereduction.html
Perhaps this might help as a guide for noisereduction?
|
| |
uneksija
Registered: Jan 2002 Posts: 44 |
Thanks ready. for the information. I'll try to take a pic of my plugs someday. I want to stress that the way I did it is _not_ a hardware modification, but simply one way to connect the SID to the outside world. |
| |
Frantic
Registered: Mar 2003 Posts: 1627 |
A picture of this modified plug at http://codebase64.org/ would be 100% nice, in order to make it available for the future. :) |
| |
gregg Account closed
Registered: Apr 2005 Posts: 56 |
<Post edited by gregg on 5/10-2007 04:03>
Just solder these two pins together.
|
| |
Devia
Registered: Oct 2004 Posts: 401 |
Uhm.. that picture shows the connector IN the computer and not the connector at the cable. On the cable it's of course mirrored.
...iirc ;-)
|
| |
Frantic
Registered: Mar 2003 Posts: 1627 |
<Post edited by Frantic on 5/10-2007 16:20>
Ok! I started a little article on this on codebase. Feel free to improve it. (Did Magervalp say something about that it was better to connect these pins with a resistor or something? I don't know shit about these matters.)
http://codebase64.org/doku.php?id=base:reduce_noise
Have a nice day! |
| |
gregg Account closed
Registered: Apr 2005 Posts: 56 |
@Devia, that's right, I just forget to mention it! :)
Shorting composite video to ground shouldn't be dangerous though. |
| |
Hate Bush
Registered: Jul 2002 Posts: 453 |
so, i was too lazy to have fun with soldering. just as i expected.
BUT it is quite effective to get rid of the noise using the noise gate. very short attack, hold and short-to-mid decay did the job beautifully for me, and they will probably do it for you as well. provided that you don't use long (B to F) release phases. and yes, don't expect to filter out the whole noise from the low frequency dominated tracks (eg. kick drum). |
| |
Frantic
Registered: Mar 2003 Posts: 1627 |
...and I just updated the Codebase page on this topic a little.
http://codebase64.org/doku.php?id=base:reduce_noise
Yep! |
| |
Devia
Registered: Oct 2004 Posts: 401 |
Hm.. This past weekend I modded my A/V plug to short SIDIN and GND.
The result was.. well.. non-audible :( ..meaning I can't hear any noise reduction improvement from that at all.
Someone should test the result of this visually by sampling the output in the two cases... I'm too lazy to do this right now :-P
|
| |
MagerValp
Registered: Dec 2001 Posts: 1056 |
Noise reduction samples:
http://www.bigmech.com/misc/c64mods/noisereduction.html
|
| |
Devia
Registered: Oct 2004 Posts: 401 |
Yes, I've seen that.. which is why I'm a bit puzzled. But maybe there's some good explanation, like the AIN pin not beeing connected on the c64 I tested or some weird shit like that.. I'll have to do some more testing ;-) |
| |
Frantic
Registered: Mar 2003 Posts: 1627 |
@Devia: I'd be curious to hear about your findings. Let us know. |
| |
Devia
Registered: Oct 2004 Posts: 401 |
It's a load of crap!
Grounding pin 26 (EXT IN) on the SID is useless. Especially grounding it in the AV plug is down right stupid, since it's already pulled down via a 100k res.
Only the original c64 schematic from the Service Manual (251138) does not have EXTIN pulled down. And NOBODY actually has a machine buildt after that schematic.
Now, there is a 100nF cap serially connected on the EXTIN line and the pull down is like I said on the EXTERNAL side and not the SID side. This means that the EXTIN pin on the SID actually is not pulled down. It seems to be internally pulled up though (4.8V on my test subject).
Grounding the pin on the SID side of the cap via a lower resistance (I tested with 330 Ohm), theoretically should produce less noise. This is however inaudible on my test speaker with the high pitch sound simply getting all the attention.
If I lift the EXTIN pin, so that it is not connected, a low frequency audible noise gets added. This goes away if I ground the now not connected pin.
Although I might be able to produce reduced noise levels by giving EXTIN a more solid (lower resistance) ground, this is useless if I do not apply other methods for eliminating the characteristic high pitch sound.
I currently have no means of sampling anything, so actual audible and visual examples cannot be given.
I am very curious as to how the fuck AlphaA managed to produce such a convincing result!?
oh.. and I tested on 2 breadbins of different PCB layout and a C64c.
|
| |
Devia
Registered: Oct 2004 Posts: 401 |
@JackAsser: Go spank your boss and make him tell us what kind of trickery he played on your c64. What kind of filter did he add to what? And what did he do to the EXTIN pin?
|
| |
Frantic
Registered: Mar 2003 Posts: 1627 |
@Devia: In case the hight pitched noise (which I understood as originating in the h-sync-business rather than EXTIN) is what is most audible on your machine, perhaps the EXTIN-line simply doesn't pick up much noise in your room and hence; no audible difference? |
| |
Devia
Registered: Oct 2004 Posts: 401 |
Yes, well.. that was sort of what I was trying to say.. It may very well be that providing a low resistance ground on the EXTIN pin will reduce noise considerably, but if it's inaudible, it's not of much value.
I still need to test this while applying other means of reducing the high pitch sound. I would like to try out whatever JackAsser did along with lda#0;sta$d011;sta$d020 to get a more complete picture of which method actually does what. Doing this on old and new c64 and c128 while sampling and documenting is a bit of work, and I'd like to avoid having to set up everything several times to test out some other means, which then would be out of combination with other methods...
I want to build a test matrix first, so to say ,-) |
| |
MagerValp
Registered: Dec 2001 Posts: 1056 |
Mogwai used a 100 Ohm resistor on EXTIN, iirc, and on that machine it made an audible difference. Unfortunately I don't remember whos computer it was, or which model. Grounding the A/V plug is just referenced in that mod article, and he does a lot of other mods that affect the result.
|
| |
WVL
Registered: Mar 2002 Posts: 886 |
Actually, while I was fixing my c64 I thought I'd check out where what is in my old 1983 model c64 (artwork 251137 rev B).
This is a c64 with ceramic vic and most other chips also socketed, and I couldnt find any schematics or mods readily available for this exact motherboard.
On this model, pin 26 (ext-in) of the sid is connected to the + side of capacitor C12, a 10nF (? ;) ceramic disk capacitor (I wonder why they put + and - on the board?). The other side of the capacitor goes directly to the AV plug.
Pin 26 is at +4.75V from ground. So grounding it directly does not sound like a great idea ;)
I'm guessing the best way to avoid any noise coming in is to hook the - side of C12 to ground. Or maybe you can hook EXT-in directly to ground via a big enough resistor. But then I'm worrying how much current the SID can handle?
Let's say 1mA max, which gives R=V/I=4.7/1e-3=4700 Ohms.
Any comments?
edit - while I'm at it -> what's up with R38 that C= needed to put 2 resistors in paralllel to get a such specific value? |
| |
SIDWAVE Account closed
Registered: Apr 2002 Posts: 2238 |
so there is still no working solution that is safe for the sid ? |
| |
Grue
Registered: Dec 2001 Posts: 146 |
Just disconnect the ext-in pin from the sid socket. It can be done quite easily when you bend the pin straight so it doesn't make contact with the socket.
You can also ground the ext-in pin FROM THE AUDIO/VIDEO CONNECTOR behind the c64. Grounding ext-in pin straight from the sid will result most likely to hardware failure.. Its safe to ground it after the capasitor.
Also adding extra filtering to +12v/+9v voltage supply have helped along the way. |
| |
SIDWAVE Account closed
Registered: Apr 2002 Posts: 2238 |
Quote: Just disconnect the ext-in pin from the sid socket. It can be done quite easily when you bend the pin straight so it doesn't make contact with the socket.
You can also ground the ext-in pin FROM THE AUDIO/VIDEO CONNECTOR behind the c64. Grounding ext-in pin straight from the sid will result most likely to hardware failure.. Its safe to ground it after the capasitor.
Also adding extra filtering to +12v/+9v voltage supply have helped along the way.
So the first you write, is the only safe way ?
(which requires no soldering) |
| |
Frantic
Registered: Mar 2003 Posts: 1627 |
Can anybody else confirm that you simply have to bend away the "ext-in" leg of the SID chip to fix the noise? If that is true (according to more than one person) I will add it to the codebase64.org article on this topic (noise reduction). |
| |
Grue
Registered: Dec 2001 Posts: 146 |
Quote: So the first you write, is the only safe way ?
(which requires no soldering)
Its safe if you bend the pin CAREFULLY, its always possible that yuu break it. Its also risky to remove chip from the socket if you dont know what youre doing...
What comes into if its helpful, it is..
I instructed alankila to do this hack and here is what he as to say about it:
"Modify your C64. Take out the SID chip and locate the third pin from top right (number 26, I think). It is the AIN (audio in) pin, and it sucks in most of the system noise, especially the display whirr. We need to ensure it is grounded or floating. The easiest fix is to make it floating: bend it slightly off, so it doesn't enter the socket on chip insertion. If you are very adventurous you could also try to improve the SID input voltage with additional capacitors and coils. It will help a little, but this modification is what counts. "
http://www.bel.fi/~alankila/c64-sw/yourchip.html |
| |
Mace
Registered: May 2002 Posts: 1799 |
Please note that Dr. 8 Bit started a thread but never replied to ANY of the suggestions. |
| |
GT Account closed
Registered: Sep 2008 Posts: 308 |
aeeben allready gave you guys the solution two years ago, and you are still talking about breaking legs on the SID chips. OMG. Just activate bit 3 on register $d417, the SID chip input should be disabled.
Replace the STA $d417 in a player with JSR to this code:
ORA #$08
STA $d417
rts
|
| |
Frantic
Registered: Mar 2003 Posts: 1627 |
Hmhöömhmhmhm...
Perhaps I shouldn't rely so blindly on AAY64, but it says:
http://unusedino.de/ec64/technical/aay/c64/sid23.htm
So.. two things...
1. What exactly is this bit in $d417 doing? Is it a flag to determine whether external input should be filtered or not (but still audible), or is it a flag to determine whether external input should be filtered or not routed into the chip at all?
2. If the info is correct, it seems to say that lda #8 sta $d417 would route the external input into the filter, and thus this code would in fact NOT disable it? (Perhaps the idea here was that if you route the signal into the filter, the *filter* will make it "silent" through filtering away some of the input noise?)
Anyone up for a clarification? |
| |
GT Account closed
Registered: Sep 2008 Posts: 308 |
With bit3 activated only, it disables the input. But setting a bandpass in $d418, the input gets filtered of course. Please give more feedback on this, I tried it on a C64 slim with 8580 chip. |
| |
Frantic
Registered: Mar 2003 Posts: 1627 |
Not allowing the musician to use the filters as he wishes makes this "trick" kind of useless for general purpose use.
It also implies that bending one of the legs of the SID is in fact a better method for general purpose use. |
| |
GT Account closed
Registered: Sep 2008 Posts: 308 |
Quote: Not allowing the musician to use the filters as he wishes makes this "trick" kind of useless for general purpose use.
It also implies that bending one of the legs of the SID is in fact a better method for general purpose use.
At least you'll get rid of most of the noise when using filter. Especially on low cutoff tunes, using heavy basses etc... The noise makes in more lively IMO. I wouldn't recommend bending pins. We're talking vintage equipment. :-) |
| |
Soren
Registered: Dec 2001 Posts: 547 |
I did bend that one leg on my sid chip, which gave me quite an improvement... along with $d011,0 and $d020,0, this should be perfect enough for me, when recording sidtunes or sid sounds :-) |
| |
GT Account closed
Registered: Sep 2008 Posts: 308 |
Quote: I did bend that one leg on my sid chip, which gave me quite an improvement... along with $d011,0 and $d020,0, this should be perfect enough for me, when recording sidtunes or sid sounds :-)
Hello Jeff.. OK, maybe I'll sacrifice one chip, though. :-) |
| |
Soren
Registered: Dec 2001 Posts: 547 |
Just do it carefully, then I can't see what could go wrong. Do ofcourse make sure that the bended leg doesn't touch any of the other legs ;-) |
| |
Devia
Registered: Oct 2004 Posts: 401 |
As I already wrote a couple of posts up (#53), lifting the EXTIN pin will add a low frequency noice.
|
| |
ready.
Registered: Feb 2003 Posts: 441 |
what about removing the capacitor connected to ext in? this will also leave the ext in floating. This might be useful for the C64 versions with soldered SID to the motherboard. |
| |
GT Account closed
Registered: Sep 2008 Posts: 308 |
Quote: As I already wrote a couple of posts up (#53), lifting the EXTIN pin will add a low frequency noice.
That's what I was afraid of. Just removing/bending a pin will leave other issues, like grounding. I get rather good quality with the audio cable I've made for the mixer, so I don't get any problems with lots of noise. So the trick by filtering the ext.in works good aswell. Remember, that with tunes not using filters, you can turn ext. in off (lda #8, sta $d417). |
| |
clonK Account closed
Registered: Aug 2008 Posts: 65 |
Anyone have any solid answer for getting rid of the noise (15550hz -> 15763hz) on an ASSY 250407 PAL?
This is a breadbin with a 6581.
Can I tie the audio-in to ground inside the machine without having to use a cap and resistor? People keep saying you need to put the cap and resistor in between, but then other people say the SID is already protected by a cap and resistor directly on the board, and then other people say that it is only needed if you have 2 SID's and that's only for the second SID! arggg.... I'm going round and round and round in circles!
The bigmech mod only is only for 8580 c64's.
I did have a video out cable where I grounded audio in, but I've since scrapped that (fell apart) and started using a new cable for LCA. I want to do the audio in->ground connection inside the 64 and I want to do it correctly.
Do I put a cap and resistor between audio in and ground for a breadbin, 6581, 1 SID noise reduction or not? Disabling the audio in for the moment will be fine.
I've trawled the Prophet64 forum and googled but still no solid answer is appearing.
Cheers
|
| |
JackAsser
Registered: Jun 2002 Posts: |
Quote: Anyone have any solid answer for getting rid of the noise (15550hz -> 15763hz) on an ASSY 250407 PAL?
This is a breadbin with a 6581.
Can I tie the audio-in to ground inside the machine without having to use a cap and resistor? People keep saying you need to put the cap and resistor in between, but then other people say the SID is already protected by a cap and resistor directly on the board, and then other people say that it is only needed if you have 2 SID's and that's only for the second SID! arggg.... I'm going round and round and round in circles!
The bigmech mod only is only for 8580 c64's.
I did have a video out cable where I grounded audio in, but I've since scrapped that (fell apart) and started using a new cable for LCA. I want to do the audio in->ground connection inside the 64 and I want to do it correctly.
Do I put a cap and resistor between audio in and ground for a breadbin, 6581, 1 SID noise reduction or not? Disabling the audio in for the moment will be fine.
I've trawled the Prophet64 forum and googled but still no solid answer is appearing.
Cheers
I can only say what I did and it worked for me. The noice you hear is the horizontal sync from the video signal. What I did was shielding the chroma and luma lines from the audio line as good as I could. I.e. by cutting with a scalpel to make better distance. Applying copper tape where needed etc. Then ofcourse only use Y/C output + audio and not the composite or RF output. IIRC I also think I copper taped the whole VIC-cage. |
| |
clonK Account closed
Registered: Aug 2008 Posts: 65 |
Quote: I can only say what I did and it worked for me. The noice you hear is the horizontal sync from the video signal. What I did was shielding the chroma and luma lines from the audio line as good as I could. I.e. by cutting with a scalpel to make better distance. Applying copper tape where needed etc. Then ofcourse only use Y/C output + audio and not the composite or RF output. IIRC I also think I copper taped the whole VIC-cage.
I don't get what you mean by "cutting with a scalpel to make better distance".
I also have a cable I built that just carries Luma and chroma.
Now I want to tap the audio out from the pcb and possibly ground the audio in.
I'm having a hard time locating the right spots on the pcb too actually... guess I should try and find schematics.
The bigmech mods don't correspond to the 6581 breadbin motherboard.
Anyone else done this? |
| |
Hermit
Registered: May 2008 Posts: 208 |
Hi
Once with one of my C64s I tried everything what is written here & there (grounding,shielding,whatsoever), but I didn't feel significant change in the SID noise level. Somehow it's still a mystery.
I even changed buffer-capacitors and injected some LC-filters to the PCB wires to SID +5V and +9V power pins, that didn't help a lot either. However the noise is clearly in connection with VIC screen signal, I have the assumption that the big noise is coming from or caused indirectly by the digital part of the chip,as no workaround on the analog part helped.
You know, it's a hybrid chip which generates analog signal out of digital counters. I think somewhere the digital signal is not clear enough, and it is felt when it's converted to analog signal...maybe I'm wrong but:
Recently I bought an oscilloscope and measured signals on C64 chips. I can tell you, in places the cock/data/address lines are so heavily used that the measured signals are far from perfect 'pulse/square' shape. They're yet enough to work well, but wire capacitance and bus-load deforms some signals significantly.
And the most surprising measure was on leg 6 of SID,which is the rude clock signal from CPU leg 39. Tapping on these legs with oscilloscope, the SID-noise changed SIGNIFICANTLY however there was no change in screen content. The noise got more even, with a bit more high frequencies IIRC. (The scope might have enough capacitance (<10pF) to alter the clock signal?!Or there's a ground loop in PCB / my scope wire works as antenna? my imagination goes too far :) Maybe the poor clock-signal from CPU to SID has to do something with the legendary noise and a clock-driver could make the situation better...maybe...
On the other hand I agree this contributes to the lively sound which we got used to, and I don't think I'd further disturb C64s in search of the hardware noise...
CHEER$
|