Bit Crusher effect

LMAO, probably was :laughing: I cant find which thread but I remember seeing it a year or two ago?
Either way it handles the ā€œnearestā€ like resampling correctly

and indeed it is: Nearest Neighbor Upsampling - #4 by steve

but it’s not really a ā€œBit Crusherā€ as it does not reduce the bits per sample. It’s just a (deliberately) bad resampler.

Haha yeah that’s what I mentioned, although I usually just reduced the bitrate on the left panel after applying the effect.

But yeah if this crushed the sample rate this way this script would be absolutely perfect for making retro genesis/GBA/DSi/XP sound effects and music :smiley:

EDIT: Just realised in October it’ll be the 10th anniversary since you wrote that script

would it be possiblee to also change the track bit depth to the bit depth it was crushed to? eg if i have 16bit audio track, and i crush it to 1bit, the audio track will also be 1bit? because currently, after crushing the track, the track is still 16bit

Short answer: No.

Audacity tracks support 16-bit integer, 24-bit integer, and 32-bit float. Audio tracks cannot be set to any other format.

Nyquist scripts are not able to change a track’s sample format.


Does that matter?

Think of it this way:
Say that you have a series of numbers:

  • 1.0, 2.0, 7.0, 3.0, 12.0

Now let’s represent those numbers with higher precision:

  • 1.000, 2.000, 7.000, 3.000, 12.000

The numbers in both lists have exactly the same values. The only difference is that the second list has more digits, which makes no difference at all to the values of the numbers.

The same is true of representing a low bit-depth (integer) waveform in a higher bit-depth (integer) format. Technically, a 4-bit linear PCM waveform can be represented exactly in 5, 6, 7, … 1000000 bits per sample linear PCM. In this example, 4-bits per sample is just the minimum number of bits per sample that are required to accurately represent the data - there is no maximum.

Does not sound like 8-bit audio (the one i was going for) and i don’t want to download external programs to get the proper effect.

What do expect 8-bit audio to sound like?

A lot of people incorrectly assume that 8-bit audio will always sound like a vintage arcade game.

so, basically, what i want is pixelated-ish stuff. and none of the stuff i’ve found (that aren’t programs other than audacity) has a good effect of what i wanted. i did find one, but it was an external program. i can show the audio example, well i can’t, because i am still too new to the site.

1 Like

how the fuck do i add .ny files

Please avoid profanities.

The easiest way to install a Nyquist plugin is with the ā€œNyquist Plugin Installerā€. See: Nyquist Plugin Installer - Audacity Manual

After installing the plugin, restart Audacity and it will then be found by Audacity.

Hello, I installed the plug-in, and enabled it. Now, where is it? I tried searching in ā€œEffectā€ but I couldn’t find it. Please help me.

It should be in the Effect menu. It may be in a sub-menu of the Effect menu. The precise location depends on how your Effect menu is sorted / grouped.

I am actually in search of a plugin that will do distorted down sampling like some other posters in this thread are. I’ve been using the ladspa plugins on windows which includes a decimator that does this kind of classic crush and down sample combo, though it has left me with no choice but to use 32 bit audacity on a 64 bit system for years now, since those plugins are contained in 32 bit dll files. I’m not even using the latest version of audacity to begin with because of the major work flow overhall things that were done to it, but that’s a story for another time. When I talk about distorted down sampling, I’m talking about what happens if you were to save an audio file at say, 8000 hz sampling rate. If you were to play it back using any good audio playback tools on good systems, it will sound smooth and be properly limited to the 4000 hz max output frequency. However, if instead you play it using a simple dac on a low speed chip like those used in Gemmy anamatronicks, electronic games like bop it, or any other cheep toy that uses low rate audio, the signel will be output raw and unfiltered, which results in a sort of square modulation at the 4000 hz max frequency since things aren’t smoothed out and all the upper harmonics of that aliasing are there to be herd. Many if not all old systems that had pcm sound of any kind, wether it’s an arcade such as mortle combat or golden axe, the gameboi advance, and even the comador amiga with it’s mod tracker files play audio in this way since you need fast hardware to be able to have accurate downsampling without a major performance drop. There are even systems such as yamaha home keyboards that are imbetween properly filtered and totally open when it comes to handling lower rate sounds, since even today with the latest keyboards for the home market, yamaha still uses a system which can’t fully filter the artafacting. It’s super apparent when you listen to the xglight pizicato strings in issolation, because those seem to be sampled at some rate that results in notes having a theretical sample rate way below 6 khz on the low part of the keyboard using that sound because the sounds are so low quality to begin with. So playing bass notes on the pizicato strings preset of most yamaha home keyboards excluding a couple from the early 2000s that had higher quality versions of some sounds results in aliasing. This high frequency aliasing becomes more apparent the lower you go with the rate, and will be nearly if not impossible to hear if you played unfiltered audio near the rates of cd sound and studio sound. Also it would be helpful to allow direct entering of the rate with all the digits, so more fine tuning to the rate can be set. This along with the possibility of having a distorted down sampler would be usefull to me since I started a project a while ago with some friends to try and figure out the true correct factory pitch/ clock speed that toys/ games should play sound at since many of the low cost handheld games and kids toys even today have some lower quality sounds, and it requires knowing aproxamately where the sample rate is for these sounds and then using a desimator, the sign modulator in gold wave’s meconize tool, etc to add what is basickly the same thing as that kind of aliasing caused by low quality down sampling and fine tuning it to get to the point where you can hear a beating between the modulation you added and the original aliasing/ modulation output from the source, so you can lock on to what the source played at and then figure out how far that is from the expected normal rate of the sound. This test of clock speeds on toys/ games is useful since a lot of them use a resister to set the speed instead of some locked time cristal to save cost, and from one unit to the next, the over all speed of everything from the gameplay to the sound can be wildly different sometimes du to the resisters not all being the exact same value. We even used it to settle debates that people have had about what the technical correct pitch for a game is, just by figuring out that the pitch some people thought was correct actually is by showing how it lines up with a number that makes sence in terms of the sampling rate. A practical example from the findings of this test follows. We took a recording of the original bop it from 1996 made using a telephone pickup coil and put it into an editor, most lilkely gold wave. Going in, we know that the rate for all the sounds on that game should be 8 khz, so we begin using the desimator, sign modulator on meconize, etc at that frequency and fine tuning it until beats between our frequency and the frequency from the source are present. slowing these beatings down as much as possible by fine tuning even more results in the fact that the unit being recorded of that particular game has an actuall sample rate of around 8150 hz, though it does fluctuate by a prety negligible amount through out because of slight voltage inconsistencies on the power source running the game. This results in a necisary playback rate change of about -1.875 percent to get normal speed or very close to it. Another example is with the bop it blast from 2005. Those as we discovered are super underclocked for the most part in general, so people assume the normal pitch to be lower than other bop it games that use the same sound set as the blast in terms of sound effects, but doing the rate test, assuming the sound is supposed to be at 8 khz since the aliasing is near there anyway with the lower pitch, does conferm that when doing the playbakc rate change from an aproxamate rate of 7556 hz to the expected 8000 hz in terms of the playback speed being changed as before, results in sound effects not including the voice which have been used in modern bop it’s since then, being the currect pitch. This was one of those things that the bop it community has had disputes over for years, and it has been prooven that most units of the bop it blast are clocked super slowly in general.

Text that is formatted in paragraphs is much easier to read than a huge block of unformatted text.

Sorry about the lack of formatting. Since I’m a blind computer user, many times if I’m writing something for a forum or message board it ends up not having any since I don’t have any easy way to see how the text looks on screen. All I end up being able to do is listen to what my screen reader software reads and that doesn’t usually include detailed formatting info unless I’m using microsoft office or something.

There’s a good free bit-crushing/down-sampling plugin from ToneBoosters called TimeMachine.
It’s available in 32-bit and 64-bit versions.

I downloaded the 64-bit version directly from ToneBoosters legacy page
ToneBoosters | Audio Plug-ins | Changelog (ā€œlegacy v3 plugins onlyā€)
rather than getting it from a third party website.

Update: I just remembered the GUI cannot be switched off on VST2 plugins in Audcaity3.
That may render the Toneboosters legacy v3 plugins useless to blind users. :frowning_face:
[With VST3 plugins the GUI can be disabled, replaced by text-only interface ].

It’s called Aliasing when reducing the sample rate like to 8KHz when 8-BIT Does Quantization Noise and the first nor latest BitCrusher.ny does not have Aliasing Added to it! even TimeMachine by ToneBoosters Has No Sharp Wave Samples

ā€œAliasingā€ occurs when the audio being resampled contains frequencies that are more than half of the new sample rate. It has nothing to do with the bit format and nothing to do with steps in the waveform. Resampling typically applies an anti-alias filter prior to changing the sample rate so as to reduce or eliminate aliasing. Resampling with this effect uses an anti-alias filter.

ā€œStepsā€ in the waveform are always present in PCM digital because there are always a limited number of possible amplitude values. This is not usually visible because audio is usually recorded with at least 16 bits per sample, which supports thousands of possible levels, so the the steps are too small to see.

how do i get it to work?