Jeroen Tel - Stranglehold (Sega Genesis, YM2612) [Cover]
Vložit
- čas přidán 7. 09. 2024
- I've wanted to do a Genesis cover of this song for a long time because I wasn't sure if it was possible. I can now confirm that it is :)
Original Version by Jeroen Tel: • Jeroen Tel - "Strangle...
Technical bits:
As for how I pulled this off, I used two main tricks: extended channel 3 and the 16xx effect. This effect changes the multipliers of an instrument patch, which, with some skill, can yield some crazy polyphony. I used a similar technique when making "Space in a Jar."
VGM: drive.google.c...
Composed in Furnace: github.com/til...
Rendered with Corrscope: github.com/cor...
That sound at 00:07 for some reasons is so satisfying
i agree
1 year ago lol
I just searched up Stranglehold Genesis cover for the first time, and yours just happens to be uploaded a few days ago :) I knew this was possible.
I love this! it's oddly relaxing, nostalgic and eerie!
amazing!!
also what's your corrscope preset, cause that's some of the best triggering i've ever seen on an oscilloscope video!
Thanks!
Well, I don't actually have one. I usually use the default settings and tweak them as needed, but in this case I didn't need to change the triggering at all!
My guess is that, since the song has only sine waves, the default triggering was enough.
@@SpinningSquareWaves interesting
Really damn cool used of the extended channel mode, also damn those sines sound clean
@jeroen tel would be proud!!
I have been waiting for this for a long time !
this is rlly impressive!! it sounds great!!
Nice Sega Genesis cover of the song!😄🤩😎👌
Wow this is crazy close to the original forgetting it’s running on that YM2612
yass
Imagine having some water drop sounds with the psg
Damn, that's a fine cover
Awesome cover :0
Fantastic
ooh nice thinking with the mult commands :)
Thanks!
I actually used a similar technique in my song "Space in a Jar," but for that one, I used it to create a single channel echo.
Channels 1 and 2 are just triggering multiple ops at the same time, right? It reminds me of what Follin did in Time Trax.
Pretty much, yeah! It's actually very similar to Time Trax except, instead of four notes split across two channels, it's eight notes split across two channels.
WHERE'S THE MAN HIMSELF?
@Jeroen Tel
This is possible on SNES soundchip?
The short answer: yes, probably
The long answer:
Since the SNES uses samples instead of FM synthesis, the limitations are going to be very different.
In theory, it would be possible to have samples of each of the sine wave chords and then play those back on one channel, leaving the other channels to just be sine waves. Using the oscilloscope of my cover as a reference, the top two Genesis channels would be played by one SNES channel, and the bottom four Genesis channels would be played by the remaining seven SNES channls.
The main drawback to this technique is that it would take up a lot of precious RAM in the audio processor, which is less than ideal but still very possible.
Additionally, it might be possible to abuse the SNES's echo filter to reduce some of the samples required by simulating some of the echo channels for the lead. Although, to be honest, I don't know enough about the limitations of the echo filter to say anything definative.
Marmelade dream in sega genesis or yamaha ma-2, or ma-3
How about yamaha ma 3
Stranglehold on the MA-3 is definitely possible.
From what I can tell, the Yamaha MA-3 (YMU762) has 32 channels FM channels. Because of this, no technical tricks would be needed as all 16 channels of the original version would fit without difficulty, making such a cover very easy. In fact, if the tricks I used on the YM2612 were to be applied on the MA-3, only a third of the available channels would be needed, or 11 channels in 2-operator mode to be precise.
Depending on the CPU driving the chip, it might even be able to play the original XM file through the sound chip's multiple PCM/ADPCM channels.
@@SpinningSquareWaves how about sine waves as frequency modulations.
@@yoelhgo02 Well, to get the MA-3 to produce the song as accurately as possible, you wouldn't want any frequency modulation. In my Genesis cover here, I don't use any modulation, as it would distort the sine waves.
If you mean having multiple sine waves on one channel, it would definitely be possible; however, it would be unnecissary, due to the large number of channels available on the MA-3. It would also be far more limited as the MA-3 (assuming it's comparable to the OPL line of chips) has more limited multipliers. Notably, it's missing 11, 13, and 15; the later of which is essential for many chords/intervals.
tl;dr - yes, easily
@@SpinningSquareWaves MA-2
Yes, about as easily as the MA-3.
From what I can tell, the only difference between them is the ADPCM on the MA-3
hello I'm back again
another question is how do you fit multiple notes into one channel?
Hi!
So, this is a really technical question.
Essentially, FM synthesis works by configuring sinewaves into different combinations of adding and modulating. One setting that's important for this is the multiplier, which can multiply a sinewave's frequency by a number between 1 and 15. By configuring the sinewaves so that they all add together, and by setting the multipliers to certain values, you can get four note chords out of one FM channel.
I'm oversimplifying a bit, but essentially, that's what I'm doing here. There are definitely limitations to this, but in this case, that technique was enough for me to get the chords I needed for this cover.
Good !!! How did you made it with FM chip?
As I mentioned in the description, there are two tricks that I'm using: the extended channel 3 and the 16xy effect. The former allows me to individually control the frequency of each operator in the third channel, which I use here for the lead. The latter allows me to abuse the multiplier settings of each FM patch to control the pitch of each operator in a more limited way. I use this technique for the chords/arpeggios found in the first two channels
Wow!!
Make one of stranglehold 2, nice cover.
Thanks!
I might do the sequel at some point, but it'd be much harder.
It has a lot more going on: more channels, half-sines, vibrato, glissandos, and probably more that I'm missing. The best I could do for half-sines is to use a technique called SSG-EG, but that also limits the notes I can play, which means little to no vibrato.
It might be possible, but it would be much more difficult than this cover already was for the YM2612 to handle.
Why does this sound quite realistic?
How do you mean?
@@SpinningSquareWaves by what i mean "realistic" is that this sounds like modern real music and it doesn't sound like something i'd hear on an average ym2612 track
Well, I suppose the one reason for that could be the shear polyphony of the song. The YM2612 usually only outputs a maximum of six voices at a time; however, with some technical trickery, this cover pushes 15 voices at once. There are probably some other subtle things contributing to the realism/modernity, but I’d guess that’s the biggest reason
@@SpinningSquareWaves very impressive!
Do you have a vgm file of this?
I do! It is now linked in the description
@@SpinningSquareWaves thank you, I wanted to play this on either my PAL or Japanese Model 1 Mega Drive
You’re welcome!
I should let you know that the cover was written for NTSC, so it will probably play slower on a PAL Mega Drive
@@SpinningSquareWaves It plays at a lower pitch on my PAL Mega Drive. Also I already have an NTSC system... (my Japanese Mega Drive)
Right. I just meant that it would sound more like the video if you used your Japanese console rather than the PAL one. Sorry, I wasn’t clear about that
That’s interesting that it plays at a lower pitch on PAL. I’ve always assumed it would play slower, since many Mega Drive games that were unoptimized for PAL had slower audio playback, such as Sonic 1
what does the 16xx effect do? btw nice cover
Thanks!
So, the 16xx effect allows you to change the multipliers of a channel's FM patch. Clever usage of multiplier values can allow for multiple notes on a single channel.
For example, if I have two operators which are both carriers (no modulation, like with algorithm 7), then I can set the first multiplier to 4 and the second to 5, which would give me a major third interval (e.g. C-3 and E-3). Or, I could set the multipliers to be 8 and 9, which would give me a major second interval (e.g. C-4 and D-4).
One downside to this is that the tuning is always slightly off. Most of the time this isn't noticeable, but an audible example of this is in the fourth arpeggio of this cover, where the last two notes of the chord are slightly out of tune.
I could probably keep talking about this technique, but that's the gist :)
Technically, the 16xx effect isn't necessary, but it's a lot more convenient than 20 different variants of one instrument :D
I believe the slightly off tuning is due to most instruments today being tuned with equal intervals. The 16xx effect may follow Pythagorean tuning instead, which is mathematically ‘perfect’. The equal intervals today are just Pythagorean tuning but with some intervals changed so they are all equal, so this is probably the cause of the slightly off tuning.