chipcrusher re-sampling vs frequency response

Quite a few users have complained about chipcrusher's peculiar 'dry' frequency response compared to what they get with other common decimator plugins. This post hopefully will explain a few things.

Lets say we bypass the bit reduction, distortion and post filtering and only concentrate on the task of downsampling the plugin's input signal. which would be say at 96kHz. and that chipcrusher's re-sampler would be at 44.1kHz, its internal maximum.

There are two important aspects to consider:

1)Typical results achievable using a vintage sampler is very different from 'your typical Bitcrusher VST'.

99% of bitcrushers/decimator plugin out there use the same tired algorithm that was posted more than 10 years ago on musicdsp.org. This method does NOT band limit the input signal prior to the downsampling, it just sample and holds using a counter... any sample!

This is not what classic samplers did. Any engineer with half a brain at least tried to filter analog audio signal so it wouldn't contain harmonics over the Nyquist frequency of the target sample rate!. If you skip this pass, you will get extra aliasing all over the spectrum.

2)Not all lowpass filters are created equal.

All versions of chipcrusher prior to v1.005(available soon) used a CPU friendly downsampling setting which - in retrospect - might not have suited everyone's taste since it was not steep enough for high frequency content.

You can see chipcrusher's default precision somewhere in this animation made using 96kHz -> 44.1kHz with a white noise as source. All the other settings will be available. We have added a new 'Precision' parameter to set the steepness/cpu use ratio you desire. BTW The first picture in the lot is from a "do not pre filter" setting, we offer 6 such settings, from 6 point spline to truncation. Aliases like crazy, but to each is own.

Making arcade cabinet impulse responses.

Here are a few pictures we took while capturing impulse responses for chipcrusher in late 2011...
Sorry for the mess, our office is a perpetual hardware dismantling lair.

GBA SP Speaker Impulse Response

Looks fun? This is what I did for each and every speaker impulse in chipcrusher's Post Processing section.

The goal is to capture not only the frequency response of the speaker itself, but also the effect of its casing and internal components: resonance, cancellations etc.

Its thus very important to make sure to properly close the unit (which can be complicated by the tight confined space) with your soldered speaker leads dandling out without changing the tonal balance of the unit.

A few carefully created test tones are then 'injected' through the leads and recorded with one or more microphones in a mostly anechoic space at a few inches from the device. 

Next the recordings are processed with custom software.

Once thats done, and we are sure the recorded IR data is valid, we need to do the inverse: reopen, unsolder,
close and make sure it works. While the microphones are set up I usually also record native console sounds (games or test code), through that same setup, for later comparison.

Luckily no unit were destroyed, and everything worked just like it did before.

(But you can imagine the stack of devices that I have in the office and in my basement)... there is a psychological condition for that, and also a TV show about it.... Rest assured I ONLY keep tech stuff. :)

Soldering '80

This is pure nerd porn, in so many ways.
And there's a whole lot more here:

https://www.youtube.com/user/paceworldwide?feature=watch

Galaxian's digital oscillator explained.

The Namco Galaxian arcade PCB generates a few different types of sounds:

1. Analog Fire sound.
2. Analog Explosion sound
3. Three Analog 'Rack Noises' (alien drones)
4. Monophonic Digital Oscillator, which does the quirky intro tune and the aliens being destroyed.

The latter being digitally controlled and generated, it makes it a perfect candidate for the chipsounds bandlimited oscillator generators. The following explains how the digital part is generated:

First, a variable speed pulse is created by using two cascaded 4-bit counters (74LS161) (B). The input clock (2H @ 1.536MHz) is divided by a variable amount according to the last 8-bit value received from the 74LS273 (A)  flip flop. This is done by the Z80 CPU (which runs the game code) when it indirectly sets the !PITCH signal and the required pitch value into the "I/0 & Ram Data Bus" by writing to a specific address in external memory.

This pulse then drives pin 13 of a 74LS393 binary counter (C) which in turn outputs a 4-bit (16 step) binary pattern: 0000, 0001,0010,0011,(...),1111  or  0,1,2,3,(...),15 in decimal.
If those 4 output bits were to be sent to a 4-bit DAC, the result would have been a pretty standard saw waveform with no volume control. But this is obviously not what the Namco engineers wanted...

In this particular design, only 3 output bits of the '393 are used (QA, QC and QD, but NOT QB).
Resistors (D) (15k, 22k, 10k and 33k) are inserted onto the bits to create a weird R-2R 'esque ladder DAC. Also some 'components' of the signal are only added into the DAC depending on two extra
CPU controlled bits (Vol1 and Vol2).  Each driving two of the four available analog gates in the 4066 (E)

If you are following so far you understand that this can generate 4 different waveforms. Once you know this, its relatively trivial using voltage divider maths to figure out what those waveforms would look like.
I used an excel sheet for these maths:

From tests on my board the Vol1 bit is always 0 during gameplay, so logically only the [0,0] and [0,1] waveforms are actually ever used in the game. My guess its that vol1 never gets to be 1 to match the volume of the analog sounds this PCB also generates.

I successfully compared the two available waveforms with actual recordings of the board:

And they were included as oscillators definitions in chipsounds 1.6 under "05 - Arcade/Galax"

Namcot163 Dual 27C020 Eprom Cart

NES eprom carts

Clockwise from top:

1)Esper Dream 2 Aratanaru Tatakai, Famicom, VRC6, mapper 26, can use regular 27C020 EPROM
2)Lagrange Point, Famicom, VRC7, mapper 85, can use regular 27C040 EPROM
3)Castlevania, NES, UNROM, mapper 2, needs modification to use 27c010/27c020 eproms
4)Super Mario Bros, NES, NROM, mapper 0, no mods needed

Currently using 1 and 3 to run tests for my next updates.
I would also need to add my Dragon Ninja NAMCOT-163 (iNES Mapper 19) in there, but I didnt eprom-ise it yet.

Address Sniffing an EPROM

New Scope

Can't wait to put it to good use. Its a major step up from my crappy PC-based USB scope. Thanks to EEVBlog for the great review.

AY8930 Initial tests!

Over concerns received about the possibility that the chips were fakes I decided to test a few of them right away  (Paypal customer is always right no? I might just get my money back if they are fakes)

So first thing I did was to place two of those chips on my Bagman arcade board, and the sound was just as one would expect a normal AY-3-8910 (the old version). So at least they are not broken CPUs or whatnot.

The concerns raised by my colleagues pointed towards  the possibility that these were just re-branded AY-3-8910 and not AY8930's, based on the date code (2008?? if 0830 is a date code I would admit its odd for a chip designed in 1990) and on the peculiar look of the Microchip logo, also that "TAIWAN Microchip" was weird and unseen before. (though this and that beg to differ).

So I undusted my AY protoboard that I used for chipsounds research, and placed a AY8930 that was previously tested OK in Bagman. Tried a few VGMX log files, (including intv_lock_n_chase.vgmx - one of my favorite games)

First interesting discovery was that the chip didn't seem to run at 3.57MHz but at half of that (1.79MHz, just ok for the right pitch of the intv register dump). I forgot that my protoboard was last populated with a YM2149F, which uses pin 26 for clock divider (input). and my jumper was already shorted. So at least if this is a rebranded chip, its a rebranded YM2149F and not a bare AY-3-8910, since the 8910 doesn't have this feature.

Started to check out the datasheet and made a little test VGMX file that plays a simple tone and pass through
A) all 32 values of volume
B) all 8 values of Duty Cycle (PWM)

Here's a lame 'instant video' of my setup and playing both Lock 'n' Chase and then, my test file on my protoboard:



The duty cycle changes are sadly NOT seamless. This is mentioned in the datasheet as well. "The new duty cycle value will take effect immediately. This may result in one period with a "random" duty cycle at the time the register is updated."

So much for maxYMiser -alike PWM without 90% cpu usage and/or IRQ reset tricks...

I haven't yet tried the advanced LFSR noise, but I'm pretty sure this batch of chip is legit. It wouldn't make sense to copy such a chip design when theres no real demand for it. Anything is possible though....

Until next time!

AY8930 sourced!

The AY8930 Enhanced Programmable Sound Generator is a very rare Microchip (who bought GI) version of the AY-3-8910. It includes better frequency range, noise, PWM and independent control of the envelope for each channel.

I was lucky to find a stash of those chips and will soon try them on my AY-3-8910/YM2149F protoboard to see if they work and experiment with them, then for a future addition to chipsounds.

As far as the interwebs are concerned it only has been used in one PC card (Covox Sound Master+) and one game (Where in Time is Carmen Sandiego). So my job will be to hack dosbox into spewing the register writes it sends to this chip and I'll make a render using my board (the usual stuff).

Since I have now too many AY8930's to care, if you are interested in upgrading your MSX, ST or whatever (untested), this might be your chance, since I will put a few of these rare babies on sale after they are tested.

JAMMA Space Invaders experiment.

33 years ago a game caught the eyes (and ears) of many people in Japan, and later everywhere else. You can read the whole story here of course, but this is not the subject of this post.

This post is about a little experiment I made in fixing and converting an original Midway Space Invaders motherboard to run in with JAMMA harness or standalone machine.

A Few notes:

-The coin input, and controls are all mapped.

-The video is now RGBS (same signal send to R, G and B, and sync tapped from somewhere else.

-The only thing that doesn't follow the JAMMA spec is the audio output, since my mod requires a separate amp (like many other games like all Nintendo stuff and Mr.DO), so its obvious that any collector has those already. The SI daughterboard requires raw 18V+center tap directly from the transformer which I don't have (using a standard PC psu here)

-Game now uses one 27128 EPROM and not 4 to 8 separate 2716 (9316B or whatever)

-No color overlays .. duh!