| |
SIDWAVE
Account closed
Registered: Apr 2002
Posts: 2238 |
How many BPM is a C64 or HardSID ?
[03:19] <rambones> any geniousses ?
[03:19] <rambones> how many bpm is a C64 raster irq ?
[03:20] <rambones> or as i this case. the music is at 51 hz... so i wanna find the sync to the beat
See. i wanna sync a sampled sid to the tracker..
So... it was sampled with HardSID which is 51 hz..
It must be around 120-130... but i am uncertain..
No clue what to google for |
|
| |
A Life in Hell
Account closed
Registered: May 2002
Posts: 204 |
This is a function, of course, of the speed in your editor, and a function of how many lines you consider to be a beat :)
So if you multiple editorSpeed x lines-per-beat, you'll get a number out that is frames per beat.
Multiply 51(50) x 60 to convert to frames per minute (3060/3000), then divide by frames per beat. |
| |
Frantic
Registered: Mar 2003
Posts: 1641 |
If we're talking PAL computer (which we are), then:
187.967033 BPM
125.3113553 BPM
93.98351648 BPM
...for the straight speeds. If you use some kind of shuffle-speed, the numbers differ of course... |
| |
Jammer
Registered: Nov 2002
Posts: 1333 |
as far as i remember, on amiga protracker - f06 regulated by vbl was around 125 BPM. in single speed steps are called every 6th frame. as beat is counted every 4th step, that gives one beat every 24th frame. 60 seconds * 50 frames / 24 frames gives more or less 125 bpm :) frantic was precise :) |
| |
Frantic
Registered: Mar 2003
Posts: 1641 |
@Jammer: Well... That 0.31 diff from 125 isn't totally insignificant, I'd say... I mean.. play for one minute and you already start to get slightly off, if synching with some Rambones-PC-stuff. ;)
...or perhaps that smiley was just a sign of you being delighted by my precise expressions. ;) |
| |
SIDWAVE
Account closed
Registered: Apr 2002
Posts: 2238 |
It almost impossible to sync a sidtune with pc samples, because of the .xxxxx deviation.
Jeroen Tel uses selfwritten special stuff, to make his sidmixes.
And ofcourse he wont give this tool to anyone :)
Even if i timestretch the sidsamples, its still impossible to hit a BPM, that doesnt go out of sync after about 1 min playing..
Cutting the samples in parts, and restarting them at spot on keypoints, makes little gaps. so also this method is fail.. |
| |
VIC
Registered: Aug 2003
Posts: 73 |
Perhaps you should use some shorter samples... Long sequences aren't ideal for remixing - use shorter loops instead... |
| |
VIC
Registered: Aug 2003
Posts: 73 |
On second thought (I'm a little rusty here, so bear with me) you could try to cut up your sid with recycle, but I guess the usefulness of this depends on your sequencer or whatever you use... |
| |
Stainless Steel
Registered: Mar 2003
Posts: 966 |
What VIC said. try cutting the sid into smaller chunks. works like a charm. You dont have to hit the actual BMP of the sid 100%, since it "resyncs" on every sequence |
| |
chatGPZ
Registered: Dec 2001
Posts: 11293 |
you could use something like melodyne ... :) |
| |
SIDWAVE
Account closed
Registered: Apr 2002
Posts: 2238 |
Like i explained:
When cutting sid sample to smaller bits, that are beat perfect..
If i have 10 samples, and they are played at
step 1-20-20-30-40 etc. f.ex
Then they sync to themselves, ok!
But the 0.xxxxxxx out of sync frequency, does
that no pc editor (tracker, multitrack editor, whatever),
that it cant sync with it.
It will sound ok to most, but to a musician who is trained to hear total sync, its clear it is buggy.
Ofcourse, if i did multitrack copy pasting of samples, and
spent like many many days on each bit, i can edit out of all problems. but this is too much work really.
Perhaps it is possible to make a sidplayer where you use 1mhz update as override to the 0.xxxxxxx - so that the sample will be synced with any modern editor ??
Like, just patch sidplay 2, to produce output at exactly 1.0 mhz !???
EDIT: i confuse myself...
To make a stable sample, with 0.0 offbeat variance, what is needed, and can it even be done ?
It dont matter that the sid plays not as on c64. it will be so close it dont make a difference, at least for 1x sid..
|
| |
Frantic
Registered: Mar 2003
Posts: 1641 |
What do you mean by stable? Do you mean a SID that runs at 125.00 BPM, in contrast to 125.3 BPM?
Some PC tools allow for 0.xx precision in BPM settings. The ones you use don't, or? |
| |
Stainless Steel
Registered: Mar 2003
Posts: 966 |
Not sure if this is a good example of the before mentioned method.
Its a last ninja remix i did 10 years ago on the PC with Fasttracker2.
It uses the original sampled sid tune chopped up to small parts and tracked along with it.
Please kindly ignore the poor mastering quality.
|
| |
Zyron
Registered: Jan 2002
Posts: 2381 |
You can probably fine-tune the SID-samples to make them sync 100% |
| |
SIDWAVE
Account closed
Registered: Apr 2002
Posts: 2238 |
Quote: What do you mean by stable? Do you mean a SID that runs at 125.00 BPM, in contrast to 125.3 BPM?
Some PC tools allow for 0.xx precision in BPM settings. The ones you use don't, or?
Yes, because i have tried this.
It is in sync for (appears to be) 1 minute.
Ok, so chopping them up, will resync.. yes...
Its just a lot of work :( |
| |
SIDWAVE
Account closed
Registered: Apr 2002
Posts: 2238 |
Quote: Not sure if this is a good example of the before mentioned method.
Its a last ninja remix i did 10 years ago on the PC with Fasttracker2.
It uses the original sampled sid tune chopped up to small parts and tracked along with it.
Please kindly ignore the poor mastering quality.
Yes :-)
I was just trying to avoid the work of chopping,
and make an automated method to get it to work.. |
| |
Frantic
Registered: Mar 2003
Posts: 1641 |
Sorry.. I didn't read the initial post properly. I missed the 51hz (is that exact?) part and the sample being from hardsid... 51hz gives you 127.5 beats per minute instead. |
| |
goto80
Registered: Jan 2002
Posts: 138 |
i get perfect sync when running a C64-tracker (on a PAL C64) with a tempo in ableton live set to 2 decimals, ie 125.31. it seems as if it is not always right though, which could be just my brain imagining, or could it have something to do with, uhm, electricity? --- (but yeah, i guess this a bit OT aswell) |
| |
Frantic
Registered: Mar 2003
Posts: 1641 |
It is kinda funny that there is something like a "standard song speed" on the C64. :)
(...even though of course not ALL tunes use this speed.) |
| |
Mace
Registered: May 2002
Posts: 1799 |
Quote:I was just trying to avoid the work of chopping PC audio programs like Sony Acid can do that work for you. I believe there are more tools (Recycle?).
Also there's the method of timestretching, available in Cubase SX 3 IIRC.
Quote:It is kinda funny that there is something like a "standard song speed" on the C64. Well, there are various speeds, of course. It depends on per how many rastercycles you cycle the music.
And then there's of course music that doesn't run on Raster IRQ. I believe Soundmonitor can be one of those... |
| |
(451)
Account closed
Registered: Mar 2012
Posts: 3 |
Quote: This is a function, of course, of the speed in your editor, and a function of how many lines you consider to be a beat :)
So if you multiple editorSpeed x lines-per-beat, you'll get a number out that is frames per beat.
Multiply 51(50) x 60 to convert to frames per minute (3060/3000), then divide by frames per beat.
I'm attempting to work this out, but I'm obviously no math genius. If anyone could tell me what I'm doing wrong alternatively posting EVERY possible SID (PAL) speeds in BPM I would be well happy!
My song tempo is 07 (in SID Wizard) and my beat is 6 lines.
7*6=42 (frames per beat)
3060/42=72,8571 BPM
Right?
When I play it back in Ableton Live with 72.86 it's not quite right, it plays a bit too slow, and it's more than just that 0.0029 difference. So I tried recalculating it with 3000 instead (71.43 BPM) and it gets better (the sync gets disturbing after about 3 bars 4/4 time instead of half a bar).
This was done with emulation (Vice) since I'm home for christmas and can't try it on the real thing, so maybe the emulation is the issue, but I can't figure out how to get the 187.967033, 125.3113553, 93.98351648 values either, so I'm thinking I'm doing something wrong in the calculation process.
Thanks in advance! |
| |
wacek
Registered: Nov 2007
Posts: 509 |
I had similar problems when porting music from other non-standard tools (jeskola buzz) into Ableton Live. The best way IMHO is setting the BPM value as a rough estimate, and then syncing manually the beat at the end of your sequence, using Live's very friendly user interface :) If you keep looking for a perfect value to enter in the clip properties, you'll never find it to my experience.
This method is much faster works best for me so far, like I said not only with 64 recordings, but also whenever you're doing a mix from different sources (analog/digital, vinyl, live music etc). |
| |
Graham
Account closed
Registered: Dec 2002
Posts: 990 |
Quoting (451)
My song tempo is 07 (in SID Wizard) and my beat is 6 lines.
7*6=42 (frames per beat)
3060/42=72,8571 BPM
Right?
PAL C64 has a base clock of 17734475 Hz, system clock is 1/18th of that. Clock cycles per frame are 312*63.
Framerate of the C64:
17734475 / (312*63*18) = 50.125 Hz
Quoting (451)When I play it back in Ableton Live with 72.86 it's not quite right, it plays a bit too slow, and it's more than just that 0.0029 difference. So I tried recalculating it with 3000 instead (71.43 BPM) and it gets better (the sync gets disturbing after about 3 bars 4/4 time instead of half a bar).
Per minute:
(17734475*60) / (312*63*18) = 3007.474 Hz
Divided by 7*6 = 42:
(17734475*60) / (312*63*18*42) = 71.607 Hz
|
| |
(451)
Account closed
Registered: Mar 2012
Posts: 3 |
Graham:
71.61 BPM (since I only can set two decimals) works REALLY well, thanks for explaining!
wackee:
Since I use Live mainly as sequencer I think I'll just fine-adjust the clips manually where needed, but that method might come in handy for future projects, thanks! |
| |
ChristopherJam
Registered: Aug 2004
Posts: 1403 |
Perhaps run off CIA instead of VIC?
Given that the system clock is (17734475/18)Hz, one fiftieth of a second passes every (17734475/18/50)=19704.972 cycles.
If you run the music routine off a CIA interrupt set for 19705 cycles, that'll get you a play rate of (17734475/18/19705) = 49.9999Hz, or 2999.996 ticks per minute.
At 24ppq, that's a speed of 124.9998bpm - that should drift by less than half a millisecond over the course of a five minute 125bpm track. |
| |
Mace
Registered: May 2002
Posts: 1799 |
Why sync the PC to the C64 and not the other way around? |
