the watmm GAS thread

[Auditor]

My sp-404 is, and will probably always be, my favorite piece of gear. Very MPC-ish excluding the sequencing (the sequencer is a joke, it's all about resampling). If you want to dip a toe into the MPC world, I would highly suggest it. It's also an amazing fx box. 

 

Although for a little bit more money you could get the real deal. 

[Guest Chesney]

Amazing how different we all are. I hated my sp404 and gave it away ha. I tried to persist with the MPC, got JJOS's and maxed it out but still, I just couldn't get on with it.

 

BUT, that's more me than the gear so anytime you read something like the above, take heed but don't let it influence your purchase.

Weirdly, I think people read things on the internet like it's gospel but in reality, most of the things you read are written by people that are so different to what you are, have different ideas or are just people with no clue spouting rubbish. There some very knowledgeable people with gold nuggest of info though too.

Reading random Internet comments have definitely put me off or hyped for me to buy something with different results.

Pinch of salt is needed.

Sorry, I went off then.

[Auditor]

No, completely understandable. Actually, using the sp to make tracks has made me bang my head against walls many times. Imo the best thing to use it for is to sample random crap into it and mangle it to oblivion. The greatest strength is its sample time. You can fart into the mic, use some delayzs, reverbs, bit reduction, and make an evolving pad out of it with resampling. 

 

Other than that, and being an fx box, it's pretty limited. 

[sweepstakes]

I think people read things on the internet like it's gospel but in reality, most of the things you read are written by people that are so different to what you are, have different ideas or are just people with no clue spouting rubbish. There some very knowledgeable people with gold nuggest of info though too.

Reading random Internet comments have definitely put me off or hyped for me to buy something with different results.

Pinch of salt is needed.

Completely agree. Also it's important to consider the person writing. Not even whether they're an idiot or a troll, but simply their taste. A workflow or sound that is poetic to them might be counter-intuitive or even obnoxious for you, or vice versa. 

 

My sp-404 is, and will probably always be, my favorite piece of gear. Very MPC-ish excluding the sequencing (the sequencer is a joke, it's all about resampling). If you want to dip a toe into the MPC world, I would highly suggest it. It's also an amazing fx box. 

 

Although for a little bit more money you could get the real deal. 

I've heard a lot of people talk up the SP-404, even over the MPC, because you can move around really fast and make tunes straight away without messing much with menus. Apparently they are pretty popular in the Seattle hip hop scene (or were a couple years ago anyway). I had an SP-202 for a little while in ~2004 and I loved the sound of the it but at the time I really wanted a sampler with good MIDI spec and its was pretty lacking. 

Nothing quite like it for getting me excited about going deep with what I already have.

This right here.

[TubularCorporation]

 

 

interesting words about MPCs in this thread.. as an aside to what I said earlier, I *do* find the MPC1000 pretty cute aesthetically, which means I'll probably at least try one someday :P but the octatrack is really hitting it for me as a super beautiful sampler and sequencer that does more than I ever thought I'd want a sampler/sequencer to do..

 

The more I've learned about the OT the more I respect it, it seems like the mid period (2000/3000) MPCs and the OT fill in almost all of each other's gaps.  An MPC1000/2000/3000 + an OT + an s3000 or Kurzweil k2000 or k2500 with the sampling addon would be a pretty incredible setup for sample based music.

 

The OT is really great, but in totally different ways. It really is apples and oranges.

 

Those two together are a lot of sampling & sequencing power. WNY/Winnie the Shit was using this setup recently, and he said they are a great combo and complement each other beautifully.

 

I personally haven't got into it personally beyond arpeggiating MPC kits with the OT (which is good fun, btw!). It's tempting to imagine the two controlling and sampling each other along with other synths, but the routing and the hemming and hawing about which machine gets sampling duty gives me a headache, lol. If I went with that setup I think any other gear would be relatively simple or menu-free synths and processors.

 

 

I've only used an OT for a couple hours in my life but from that I think I'd use it for drum programming and polyphonic MIDI parts, and do everything else on the OT.

 

Incidentally, on your OT how do you deal with clicks on pickup machines?  I was able to make it click at the loop point with nothing plugged in to the inputs at all (not even cables) and the record gate threshold set really high to catch any self-noise.  Definitely seems to be a bug, and I could replicate it on the OT I used - if there was no input signal it wouldn't click, but if I kept recording audio into the pickup machine (using the manual's recommended settings, with or without the gate on) until I got a click and then unplugged everything from the inputs and recorded an empty loop of the same length, it would consistently insert a little spike at the end of the buffer. I even saved the audio from the buffer for both steps of that process as evidence that it wasn't just user error.  Have you found a way to dodge this or has it not been a problem for you? 

[TubularCorporation]

 

Personally I think a lot of it is down to 24ppq resolution still being coarse enough that even with quantize off there's still just enough quantization happening to affect the feel, but this is all getting in to mpc-forums superstition territory.

I like this theory, it's a no-nonsense explanation of why the MPC's timing feels more precise than forgiving but you can still get bang on with enough perseverance. edit: Whereas with the Octatrack it's always, always off for me when I real-time record. Of course, it's probably more my shitty timing than anything else.

 

 

Well, if that Gearslutz thread is right and alternate 16th notes are consistently something like half a millisecond behind the beat then I'm throwing my hat in with the "90s MPCs impart a magical 'swing' to everything" camp after all. Mostly, though, that thread confirmed that the timing on the MPC2000 clock is tighter than any DAW Ive used, so in theory there's still a strong case for slaving the old MPC to the DAW via SMPTE and then slaving any other MIDI gear to the MPC's clock output.  Even though I'm too lazy to do it that way unless I'm already using the MPC anyhow.  One day I dream of actually having the table space to set up more than two or three pieces of gear at once.

 

True, that's ostensibly below the threshold of human perception but so is tape flutter on a high end machine and it still imparts something into the sound that you don't really perceive but you definitely feel (as evidenced by that col-but-ridiculous audiophile oriented process from a few years back that isolates the bias tone on a high resolution digital recording of a master tape, and compensates for the inaudible pitch variations that come from resonance in the tape itself from the friction of passing over the heads during recording and playback - you don't actually hear it when it's there but when it's compensated for there's a noticeable change.  Kind of  a tangent I guess, but I find that whole thing really cool from a technical standpoint, even though it's mostly just used to sell "better than the master" 1/4" reissues of Fleetwood Mac albums to rich audiophile types for $500.

Edited May 22, 2017 by RSP

[TubularCorporation]

My sp-404 is, and will probably always be, my favorite piece of gear. Very MPC-ish excluding the sequencing (the sequencer is a joke, it's all about resampling). If you want to dip a toe into the MPC world, I would highly suggest it. It's also an amazing fx box. 

 

Although for a little bit more money you could get the real deal. 

 

Never used a 404, but Limpyloo sent me his broken 303 a couple years ago and I fixed it up and use it constantly.  I don't actually build tracks on it much but I use it for sampling and resampling and just running stuff through the effects all the time, t really has a great raw sound to it, especially at the middle quality setting. The PCB actually had cracks in it that I had to sort of bypass though, so it's too fragile for me to play out with it or anything (not that I've been playing out much lately, although two unsolicited show offers so far this month so maybe that's turning around).

[TubularCorporation]

The MPC posts here are almost poetry, lovely to read and to think about. This is why I like watmm gear discussions; it's less LOOK AT THIS COOL SYNTH and more about workflow, the secret corners people have found on their gear, etc. Nothing quite like it for getting me excited about going deep with what I already have.

 

I'd like to say that this is what made me finally register here - it's what made me lurk for a year or two for sure - but the sad truth is that I registered to post in the "Post a Terrible Track" thread.

 

 

EDIT: pardon the post flood there.

Edited May 22, 2017 by RSP

[xox]

Well, if that Gearslutz thread is right and alternate 16th notes are consistently something like half a millisecond behind the beat then I'm throwing my hat in with the "90s MPCs impart a magical 'swing' to everything" camp after all. Mostly, though, that thread confirmed that the timing on the MPC2000 clock is tighter than any DAW Ive used, so in theory there's still a strong case for slaving the old MPC to the DAW via SMPTE and then slaving any other MIDI gear to the MPC's clock output.

 

Daw's clock doesnt drift/jitter midi messages. It only jitters midi signal if you send midi out over interface to a hw synth but when working in the box it's 100% precise.

[TubularCorporation]

 

Well, if that Gearslutz thread is right and alternate 16th notes are consistently something like half a millisecond behind the beat then I'm throwing my hat in with the "90s MPCs impart a magical 'swing' to everything" camp after all. Mostly, though, that thread confirmed that the timing on the MPC2000 clock is tighter than any DAW Ive used, so in theory there's still a strong case for slaving the old MPC to the DAW via SMPTE and then slaving any other MIDI gear to the MPC's clock output.

Daw's clock doesnt drift/jitter midi messages. It only jitters midi signal if you send midi out over interface to a hw synth but when working in the box it's 100% precise.

 

 

If you're rendering your mix of-line that's true, but when you're playing back in real time the MIDI timing can get pretty rough simply because of how a general purpose OS works. I's easy to test, just set up a few bars of quantized short MIDI notes playing a soft synth with very sharp attacks (like a hi-hat sample or a burst of white noise or something), route it to an output on your audio interface, loop that back out of the box (with an actual cable) and record it onto a new track, then line up the leading edge of the first note of recorded audio with the leading edge of its matching MIDI note in the DAW to account for interface latency, and look closely at how well the rest of the audio lines up with the MIDI track.  Maybe you'll be lucky and they'll all line up dead on but that's never happened for me, there's always some jitter between the MIDI sequence and the audio recording of the MIDI sequence. Of course, if you're 100% ITB unless you're using Pro Tools there's no reason I can think of NOT to render your final mix off-line so it's kind of moot (and I think Pro Tools finally added rendering in v11 but I never upgraded past 10 because I really don't like it and only own it because I needed it for work for a couple years).  It's also not nearly as bad as it was with old, slower computers.  That's supposedly part of why a lot of older producers stuck to their Atari STs well into the late 90s, they didn't have a multitasking operating system so there were no OS-induced timing irregularities, but that was before my time, I was just a kid with a tape recorder back then.

 

Anyhow, these days I use almost all hardware and only use the DAW clock to sync arpeggiators and drum machines and things when I'm recording in multiple takes.  I don't actually bother with the SMPTE business because for what I do I don't need to worry about clock jitter anyway, it's all pretty loose anyhow.  But if I was producing with just samples in the MPC and then recording them into the DAW in multiple passes I'd definitely use SMPTE over Midi clock.

 

EDIT: I don't think DAW cock jitter is much of an issue anymore, though.  I don't really know, but I get the impression that general purpose computers have much better clocks now; back in the golden age of MIDI (90s) they were REALLY cheap and drifted a lot, and all of the internal timing of everything the computer did was based on that clock, so anything happening in real time could only be as accurate as that (and in practice was a lot less accurate because of how multitasking operating systems handle requests for CPU cycles etc. combined with the much, much lower processor and bus speeds and other factors I only vaguely understand).  But I definitely caught enough of the tail end of that to remember it.  Even computers from the mid 2000s have internal clocks that are loose enough that the OS system clock (which, if you aren't syncing it to something external like we all do now that everyone is on line all the time, was derived from the computer's clock, basically just a cheap crystal oscillator) that they would drift wildly and need to be reset regularly.  Even now, the coputer I record on isn't connected to the Internet at all and it loses maybe 6 hours a year.

Edited May 22, 2017 by RSP

[Auditor]

 

 DAW cock jitter 

 

*Album title

[Plarkreluke Banloboh]

I'm getting an Octatrack soon so can answer Gas queries on functionality and maybe do video request tutorials, I'm super hyped for this!

[sweepstakes]

RSP I don't actually use pickup machines :) I love live sampling but haven't really got into loopers; I prefer to do it as part of a sequence. I guess this is because I was introduced to the concept by Jeskola Buzz's sample table recorder module which works very similarly.

 

On computer clocks - the network clock can actually go backwards (e.g. in Autumn when DST ends) so it is still useful to have an ever-increasing or so-called monotonic clock. Intel systems tend to implement this but some architectures (e.g. ARM at least in Raspberry Pi) don't bother anymore which has some weird implications.

[Plarkreluke Banloboh]

Has anyone ever sold time code jitter?

[TubularCorporation]

 

 

 DAW cock jitter 

 

*Album title

 

 

My fingers are apparently more witty than my brain.

 

I'm getting an Octatrack soon so can answer Gas queries on functionality and maybe do video request tutorials, I'm super hyped for this!

 

 

I have to admit, the discussion that was, I think, earlier in this very thread (too lazy to search for it) about my skepticism about the OT made me reexamine them since, realize that almost all of my concerns had been addressed by the 1.2 update years ago, and I ended up blowing most of the unspent remains of all the extra cash I earned this winter on a unexpectedly high paying editing contract, and for the most part it's really, really nice.  I'd already decided that the last big block between the way I've been working right now and being able to actually perform live without a computer (even if I liked performing with computers my laptop is so old and fussy I'd still have to make a pretty big investment to get one that I could rely on) was a sampler with the kind of realtime sampling and sample-manipulation capability that the OT has and the MPC2000xl definitely doesn't have. That's pretty much why I don't have any interest in getting any new stuff now, it feels like I could spend the next decade just sampling what I have and making music with an MPC/OT combo (or even just the OT, although I don't think I can give up the things I love about the MPC workflow, especially for drum programming, even if I only use it at home and resample the stuff I program on it into the OT for playing out) and not run out of ideas.  It clicked with me right away, although not in the same way the MPC did. I get why some people have trouble grasping it at first but for me, as soon as I actually used it I realized that more than anything it's like a powerful hardware version of an oldschool tracker, especially the sequencing.  When I was reading up on it and deciding if I really wanted to take the plunge - I've literally never spent more than $800 on a piece of gear before in my life (the $800 was for the used rack server I've  used for all of my recording since 2013 and even that was an unusually big purchase) - it sounded like it would have a really alien workflow that took a lot of time to adjust to, but then when I actually had it in my hands it felt totally natural, like a 21st century version of Impulse Tracker created in hardware.  I'm back to being pretty much broke after living expenses for at least the rest of this year but it was absolutely worth it. I'll probably be selling of some stuff over the summer that I just don't need to have around anymore (anyone need an almost unused EHX Hog II with all of the accessories and a hard case? I got it below wholesale a few years ago so I could offer a pretty low price on it; the ebay buyer never paid).

 

I think you're going to love it when you get yours.

 

I do still wish it had another pair or two of outputs, though.

[TubularCorporation]

RSP I don't actually use pickup machines :) I love live sampling but haven't really got into loopers; I prefer to do it as part of a sequence. I guess this is because I was introduced to the concept by Jeskola Buzz's sample table recorder module which works very similarly.

 

On computer clocks - the network clock can actually go backwards (e.g. in Autumn when DST ends) so it is still useful to have an ever-increasing or so-called monotonic clock. Intel systems tend to implement this but some architectures (e.g. ARM at least in Raspberry Pi) don't bother anymore which has some weird implications.

 

Oh hey, you posted this while I was typing the last wall of text glad to see I'm not the only one who has noticed how tracker-like the OT is.  Makes sense, with Elektron's history with the SIDstation and all.

 

My other post about jitter in DAWs is a mess because there were at least three different things I was trying to compare that are all called clocks, but what I was getting at is that historically the MOTHERBOARD clock tended to not be super accurate simply because manufacturers typically used the lowest priced crystal oscillators they could get away with and that cascaded down into the rest of the system, and one of the side effects of that (plus, much more importantly, how multitasking operating systems prioritize CPU cycle requests) was that the system clock was pretty far off from any kind of international reference and would drift a fair amount.  Not quite the difference between an atomic clock and the battery powered alarm clock you'd buy at a drugstore but same concept.  Anyway, digging around a bit now it looks like the technology has changed even more than I realized and that's even less of an issue now than I thought although I've still been able to clearly demonstrate MIDI timing innacuracies on my system (hardware is from around 2011) just out of curiosity - I don't actually care about it in practical use so much as I find it kind of abstractly fascinating.  It sounds like you know a lot more about this than I do, though, and I'm probably wrong about a lot of it.

Edited May 22, 2017 by RSP

[Chris Toffer]

What was wrong with the SP-303 and how'd you fix it? I've got a broken one myself - coming up with Error 8. Pretty sure there's a broken component. Opened it up and did a bit of testing but couldn't find where it's messing up

[TubularCorporation]

Two of t

 

What was wrong with the SP-303 and how'd you fix it? I've got a broken one myself - coming up with Error 8. Pretty sure there's a broken component. Opened it up and did a bit of testing but couldn't find where it's messing up

he three knobs had been smashed in and instead of breaking they had actually pulled chuks out of the PCB where the solder pads were, so I had to scrape off the insulation o some of the traces and solder from the pots to the scraped parts.  Wouldn't be too bad except the traces are pretty small so there wasn't much surface area to solder to, so it would probably be pretty easy to break it again if you weren't careful.  But it works just fine now, I just don't want to push my luck.

 

 

 

 

EDIT: you may have already found it, but I just dug up a thread on sp-forums about that error.  Doesn't sound promising:

 

When the screen displays "Err" it is stating there is a problem (0-14) which in case you had error "8". Here's what I've found:

(Solder and pattern defection around the DSP IC and EFFECT RAM IC. Solder and pattern defection between the DSPIC and CPU IC. Device defection of DSP IC or EFFECT RAM IC.)

Worst case scenario an actual component is bad, but from what it sounds like after you dropped it and it worked (momentarily) you probably somehow wedged a broken connection back together. It could be that some soldered component is not stuck on the board anymore after prolonged use and needs to be resoldered or there is a broken trace between two points. I would immediately open it up and check thoroughly looking for anything that looks loose or tarnished. Put up some pictures of the board too so it would help us help you further. Hope you find something! Good

 

Edited May 22, 2017 by RSP

[modey]

Incidentally, on your OT how do you deal with clicks on pickup machines?  I was able to make it click at the loop point with nothing plugged in to the inputs at all (not even cables) and the record gate threshold set really high to catch any self-noise.  Definitely seems to be a bug, and I could replicate it on the OT I used - if there was no input signal it wouldn't click, but if I kept recording audio into the pickup machine (using the manual's recommended settings, with or without the gate on) until I got a click and then unplugged everything from the inputs and recorded an empty loop of the same length, it would consistently insert a little spike at the end of the buffer. I even saved the audio from the buffer for both steps of that process as evidence that it wasn't just user error.  Have you found a way to dodge this or has it not been a problem for you? 

 

Could it be something to do with the FIN/FOUT settings? Having said that, I had a quick look on elektronauts and it seems like a common problem, even with appropriate FIN/FOUT. I haven't come across it since I haven't really tried looping with it yet—my initial attempt was very confusing and resulted in me saying "fuck it" and going back to using track record trigs lol.. currently the only thing I've found confusing about the OT.

[TubularCorporation]

 

Incidentally, on your OT how do you deal with clicks on pickup machines?  I was able to make it click at the loop point with nothing plugged in to the inputs at all (not even cables) and the record gate threshold set really high to catch any self-noise.  Definitely seems to be a bug, and I could replicate it on the OT I used - if there was no input signal it wouldn't click, but if I kept recording audio into the pickup machine (using the manual's recommended settings, with or without the gate on) until I got a click and then unplugged everything from the inputs and recorded an empty loop of the same length, it would consistently insert a little spike at the end of the buffer. I even saved the audio from the buffer for both steps of that process as evidence that it wasn't just user error.  Have you found a way to dodge this or has it not been a problem for you? 

 

Could it be something to do with the FIN/FOUT settings? Having said that, I had a quick look on elektronauts and it seems like a common problem, even with appropriate FIN/FOUT. I haven't come across it since I haven't really tried looping with it yet—my initial attempt was very confusing and resulted in me saying "fuck it" and going back to using track record trigs lol.. currently the only thing I've found confusing about the OT.

 

 

Nah, it's definitely some weird bug.  I think a few samples of junk data might get left in the record buffer at the loop point or something.  It only seems to happen if I record a new loop into a buffer that already has audio in it, and like other people over there have reported it stops happening for a while if you reload the project (also sometimes goes away on its own). My only new contribution to the mystery is that I'm confident I've conclusively shown that it isn't a regular clock that you would get from a bad loop point, sinceI can make it happen when I'm recording literal digital black into the pickup machine (nothing plugged into the inputs, gate pushed up all the way to catch and self noise - only thing I didn't try was recording without any inputs selected at all. Later this week I'm going to put together a little case file kind of thing with the audio files, some screenshots of the clicks in an editor, and a detailed step by step description of how I produced them, and post it over on elektronauts.  Maybe if enough people care, can replicate it and open support tickets Elektron will address it while they're still supporting the OT, although with something like 2 years since the last update I'm not holding my breath. it's not anything like a deal breaking issue for me, I rarely use pickup machines either so far, but if they didn't click so much maybe I would - on paper they act the way I've always wished other loopers did, and I find them really easy to use bt not that relevant to the kind of stuff I'm working on right now.

 

The manual doesn't explain them that clearly but if you've ever used an Akai Headrush for looping they're not much different (other than being able to quantize the loop length if you want)

Edited May 23, 2017 by RSP

[sweepstakes]

It sounds like you know a lot more about this than I do, though, and I'm probably wrong about a lot of it.

Nah I have just been working on this MIDI project for a couple years now and I had to learn a little about those clocks.

 

https://stackoverflow.com/questions/3523442/difference-between-clock-realtime-and-clock-monotonic 

https://www.softwariness.com/articles/monotonic-clocks-windows-and-posix/

 

It is kinda cool to think about at a low level, though. Like, yeah, there's a little crystal in the CPU that provides this super useful feature for doing proper timing. TIME CRYSTAL.

 

 

 

 

 

[Guest]

 

Mmm, some gas problems with this one atm looking for an on the go laptop-alternative;

 

1200 euro's though, if it comes down to 1000 in the future I can psychologically deal with it better. 

 

I've heard from an ex-Akai friend (almost everyone I knew there has left for better paying jobs in the last two years) who still follows that stuff pretty closely that you're still pretty much tied to the computer for most of the advanced features,  it that helps you resist. 

 

It's a lot more money but that new Pioneer SP-16 sampler thing looks really good to me.

 

 

Thanks for the tip on the pioneer, I will look into it, search for comparisons.

 

I have a broken, haunted Korg esx here which has sparked the hardware sampling interest & atm the mpc live is hitting all the right spots for taking apart the dug up crates ;-) . 

[TubularCorporation]

 

It sounds like you know a lot more about this than I do, though, and I'm probably wrong about a lot of it.

Nah I have just been working on this MIDI project for a couple years now and I had to learn a little about those clocks.

 

https://stackoverflow.com/questions/3523442/difference-between-clock-realtime-and-clock-monotonic 

https://www.softwariness.com/articles/monotonic-clocks-windows-and-posix/

 

It is kinda cool to think about at a low level, though. Like, yeah, there's a little crystal in the CPU that provides this super useful feature for doing proper timing. TIME CRYSTAL.

 

 

 

 

 

 

 

 

That reminds me of some old William Gibson interview I read someplace where he talks about how when he wrote Neuromancer he had never even seen a computer in person and when he finally got one years later he was shocked at how primitive it was, he thought it was going to "run on crystals or something."

 

EDIT: take the second hand ex-Akai employee stuff with a grain of salt, too.  There are a lot of them in this town and most of the ones I know are a bit disgruntled so they aren't exactly unbiased.

Edited May 23, 2017 by RSP

[peace 7]

More GAS potential fuel.....:

 

Roland announcing FOUR DRUM MACHINES this week.

 

DA FUQ?!  ...and 10 other shits.  Maybe "drum machines" are just synth drum kits...  Or could it be the ever-wanted analog 808/909/303 Roland remakes?!  ReBirth analog hardware version: $150, somehow.  Everyone in the world's asshole explodes from pleasant surprise.

[sweepstakes]

peace, my dude, you are like the Rubin Farr of GAS

 

edit: <3 u

Edited May 24, 2017 by sweepstakes