solarbird: (molly-thats-not-good-green)

I’m finally getting back to working on a new gateway/router server and I’m basically setting up this old-school sort of DMZ, with the rest of our servers hanging off one card, and our internal LAN/DHCP/NAT side hanging off the other. (Using ISC, which Debian seems to like.) And all of that seems to be right from the new server’s perspective, which is yay!

Except there’s no packet forwarding from the DHCP side even though it’s enabled and I’m sure I enabled it and yes the kernel thinks its enabled but it isn’t happening.

Any ideas where to start?

Mirrored from Crime and the Blog of Evil. Come check out our music at:
Bandcamp (full album streaming) | Videos | iTunes | Amazon | CD Baby

solarbird: (tracer)

I’ve updated the Kitting Out Cheap guide to building your home recording environment on a low budget. New additions include short commentary on kit microphones, updated interface information, and pointing out that hey, guess what – USB chipsets matter!

So, hiya, Home Recording panel attendees – here y’go!

Mirrored from Crime and the Blog of Evil. Come check out our music at:
Bandcamp (full album streaming) | Videos | iTunes | Amazon | CD Baby

solarbird: (made her from parts)
I made another thing! A magnetic (electric) Irish bouzouki pickup, the kind used on electric guitars, but with some extra features. I retconned a project name for it (Kohaku A-B-C) because it has three primary modes – “single coil,” “humbucker,” and “electric/acoustic mix,” which to be honest is kind of two modes depending upon the single coil vs. humbucker setting, but Kohaku A-B-C-D is too long and didn’t sound enough like a JDF anti-Godzilla plan.

Sounds!


Originally I was just going to do a basic wire-up with a level knob, but then I found out that humbucker pickups gave you separate access to both coils, and so can be used as a single-coil pickup as well as in humbucker mode. So that meant adding a switch. Which means it’s probably worth doing a tone knob while I’m at it. And while I was thinking about that I realised there was no reason I couldn’t also build in a tap to the extant acoustic pickup, which means I should also add a little mixer to control relative levels of the two pickups.

So I did! That was a good idea.

Details omg )
solarbird: (pingsearch)

In an audio device with a tone knob, and an on/off switch by that tone knob, which “ON” makes more UI sense?

What does "ON" mean?

  • [ ] ON: TONE KNOB is on, knob has effect on sound
  • [ ] ON: BYPASS is on, knob has NO effect on sound.

CLICK HERE TO ANSWER

This is an honest question. I mean, I’m going to wire it up so that “ON” means “tone knob on,” because that’s how I roll, but I see this go the other way a lot.

(Also in this case “tone knob” means “low pass filter” which means “removes high frequencies” because AND THERE YOU GO. See? SEE?)

Mirrored from Crime and the Blog of Evil. Come check out our music at:
Bandcamp (full album streaming) | Videos | iTunes | Amazon | CD Baby

solarbird: (made her from parts)

Literally the first signals off the electric test harness for Kokahu.

Recording is from a couple of days ago. There have been meaningful additional developments since this, but this is what I’m posting now.

How many electric Irish bouzoukis are there in North America? One more than before last week.

More to come.

Mirrored from Crime and the Blog of Evil. Come check out our music at:
Bandcamp (full album streaming) | Videos | iTunes | Amazon | CD Baby

solarbird: (made her from parts)

Remember a long time ago – like, a little over three years ago – I built a ribbon microphone? I had all kinds of problems chasing radio interference ghosts and stuff, it was strange and messy but came out with a neat sound in the end.

Except… even after fixing the RF problems, it was kind of noisy. Not unusably so, not for direct-miking, which is how I’ve used it, but still… kinda noisy. Noisier than it should’ve been.

I rediscovered this when trying to use it in a “mid-side” type mic setup with the new RK-47, which I was doing just to see how that would work. (Tony of Vixy & Tony has been after me to try that for a while.) And because it involves playing towards a figure-eight microphone from the side – the point of least sensitivity – it required enough extra gain that the noisiness became a problem.

Since the special preamp (also a kit) was the entry point for the RF noise, and since said amp works with dynamic microphones, I tested that for noise, using an SM-58 as input. Dead silent, cranked all the way up. Result: it wasn’t the microphone preamp’s fault.

Then I remembered how the RK-47/990B build manual talked so much about making damn sure you had no solder rosin or finger oils at the high-impedence connection points in the circuit, and to just scrub those connections with isopropyl alcohol. So I took apart the ribbon microphone, redid those solder connections while I was in there, and then scrubbed the hell out of them. A downright confusing amount of old solder rosin came up when I did so.

Result? Problem sorted. Huge drop in noise. There’s still a little at the high end at probably more gain than I even need here – this may be an “only elves can hear this” moment, at least in part – but a little -3db cut starting at 14-15kHz sorts it. It might not even be true noise, it could be something like air movement – ribbons will pick that up in ways nothing else will, and I didn’t turn off the HVAC, etc.

So, yeah! Turns out that finicky bit about solder rosin and flux is real important, kind of generically, at high impedance. Good to know. (And is why I’m posting this, and why I’ll link to it from the old microphone buildout writeups.)

I still have not the vaguest idea why so much rosin ended up on those connection points. Seriously, it’s weird. That was my old stock of Radio Shack silver solder, which I’d had since I Don’t Even Know When, and not the BenzOMatic solder that gave me so many problems. I never noticed it doing that before – but then again, I wasn’t really looking. ¯\(ツ)/¯

Anyway, have a test recording I made at 1:30 yesterday morning trying out that configuration, with the Micparts RK-47/990B kit mic being used as the “mid” and the above-mentioned Austin OTA-1 ribbon microphone as the “side.” It’s intermixed with a recording made simultaneously using a pair of M-Audio Novas in a spread X-Y configuration. Both versions are mixed directly to mono, rather than spread-stereo, which is not what you usually do, but does allow maximum left-right placement in a mix.

(This may be called “T” rather than “mid-side” since mid-side includes a kind of subtractive mixing not used here? I dunno. But this is, again, just a straight mix to mono.)

The recording starts with the RK47/990B plus OTA 1 pair, then switches back and forth between that and the Novas. Remember: all of this is mono.

Kinda neat, eh?

Mirrored from Crime and the Blog of Evil. Come check out our music at:
Bandcamp (full album streaming) | Videos | iTunes | Amazon | CD Baby

solarbird: (made her from parts)

Build reports are nice enough. (I wrote up a little errata post yesterday, by the by.) But the real question, of course, is how does the RK-47/990B kit mic sound?

Early impressions are surprisingly good. Even with only the single microphone, there’s a sense of presence and space – even with a purely mono signal path – that I normally have to dual-mic to attain. Also, it has tremendous precision – this is a mic capable of great subtlety. And the amount of gain built into the microphone itself is crazy – this is one spicy meatball of a microphone. That’s something you won’t hear in recordings, but it results in a lower noise floor, which is always good.

Let’s start with some unsubtle differences, ones that’ll show up on laptop speakers. Because while I’ve never liked the MXL-990, they sell a zillion of ’em, and we should make a couple of direct comparisons.

Here’s a snippet of chords from “Lukey,” alternating between the MXL-990 (unaltered factory) and the RK-47/990B. It starts with the MXL-990, then transitions in-song to the RK-47, then back and forth. It ends with the RK-47. It’s a pure mono signal path until prepped for uploading to Soundcloud.

And here’s a short melody, on zouk – again, starting on MXL-990 (factory stock), then RK-47, then back to MXL-990. The last phrase is repeated to allow us to end on the RK-47; also, I wanted that ending bit to be presented on both microphones. The glissando really highlights some of the differences.

But that’s shooting fish in a barrel, as it were. The MXL-990, while popular, is not a good microphone. We should do comparisons to microphones I actually like – let’s say, the M-Audio Nova. At about twice the price of the MXL-990, it’s still a cheap microphone, just one I consider entry-level competent. But it has issues – not the least of which being it’s kind of a noisy beast as these categories of microphones go.

So let’s take the easy swing – here’s a sharply boosted noise level comparison of the RK-47 to the M-Audio Nova, at equivalent gain levels. This is not the noise you’d actually hear; I recorded a silent room at gain appropriate on each microphone for instrument recording, then cranked that recording up 32db for easy noise levels comparison.

Unfortunately, this really requires headphones, because it’s RK-47 on left, Nova on right:

NOT SUBTLE. But also, an easy shot. The Nova is noisy and everybody knows it. There are some mods out there to improve that, but they change the sound a bit in ways I don’t like, so I work with it.

comparison of waveformsSo let’s dig down a bit. Pictured here is a snippet of waveform from a bit of music played, in mono, over my studio monitors, into identically positioned microphones relative to those speakers. These two recordings were made simultaneously.

You’ll note in this highly-zoomed-in render how the RK47 waveform remains clear and unmuddled in these extremely rapid changes, while the Nova’s blurs into a bit of a mush. That’s the sort of thing I’m talking about, and also, the sort of thing you can hear in these very short snippets of horns from a jazz track. This comparison requires headphones, possibly good ones:

They’re short because they must be uncompressed for best comparison – sorry about that – but listen to them a few times and compare. Note how the edges of detail – bits which add flavour – are blurred in the Nova, but retained in the RK-47. Neat, eh?

That out of the way, let’s step up a level in comparator microphones. Oktava 012s are considered very good affordable microphones, particularly strong in their price ranges, and street for a new 012 and one pickup is comparable to the cost of this kit. With a second head (to add a second pickup pattern, as this mic has), it’s a bit more. They’re small-can capacitor instrument mics, rather than large-can, but we’re doing instrument recording, so that’s fair. The components inside – particularly older ones picked up used – can be a bit dodgy, but the design is great and the pickups are great, and you can upgrade the iffy capacitors and the suspect transistor if necessary. I have, of course, done this with mine.

Here’s the intro to “King of Elfland’s Daughter,” on the Oktava 012 (upgraded components) and the RK-47/990B kit. This recording repeats phrases, with the Oktava 012 first, then the RK-47/990B. Pure mono signal path, identical recording setup made within a few minutes of each other, but not simultaneously, as you can’t put two microphones in exactly the same place and I wanted the most equal comparison I could, modulo performance limitations. This probably also requires headphones, as the 012 is a pretty darned precise microphone itself:

44.1khz/16-bit uncompressed WAV file version here.

Once again, I’m finding that the RK-47 has a real staging advantage. There’s a sense of in-the-room presence with the RK-47 that I can make happen by dual-miking with my other microphones and mixing down, but not directly in mono.

Now, I don’t want to leave the impression that it is BEST AT ALL THINGS, because it’s not. These aren’t the only recordings I made – they’re just ones that show differences best. The first example I found was mandolin – the Nova likes my mandolin better than the RK-47 does. The specific response behaviour and foibles of the Nova work in its favour; a single RK-47 may have more presence and precision than a single Nova, but the Nova recording sounded more musical just the same. I’m sure there will be other examples as well.

In the end, I think this will probably become a heavy-use microphone in my kit. It may even become my go-to mic on the zouk – I need to do some stereo and multi-distance comparisons before I will know that for sure, but it’s looking very good. I also like what it does with piccolo and flute. I haven’t done any playing around with fiddle or drums, and one thing I want to play with is a two-mic setup with the ribbon kit mic I built, to see how those behave together – it’s a mic placement technique I’ve wanted to try for a while, but have never got round to testing. Now is probably the time.

I kind of wish I’d ordered the matched-pair version of this microphone kit. But it would’ve cost twice as much and I couldn’t know in advance I’d like it this much, so.

Mirrored from Crime and the Blog of Evil. Come check out our music at:
Bandcamp (full album streaming) | Videos | iTunes | Amazon | CD Baby

solarbird: (made her from parts)

A bit of errata on the micparts rk-47/990B kit microphone build report yesterday:

First, I didn’t mention that the board kit includes extra capacitors, to be used either for different capsules than I chose, or to tailor the response curve of the microphone. I went for flattest response, but I could’ve had more or less high-end responsiveness by choosing which included tone capacitors to use.

Second, I said that the “omni” switch setting probably wasn’t really going to be actually “omni,” but was more likely to be figure-eight or the like. This was in response to both the capsule design, grille design, and older pictures of the circuit board which appeared to indicate that. In testing, I discovered that I was mistaken – as you can see here, levels as seen in waveforms made by holding a tone generator at the four cardinal directions remain constant. But read on, after the graphs, because there’s a catch:


rk-47 set to omni, against tone generator at 30cm, front, right, rear, left positions

And here’s the same test, with the mic set to cardioid:


rk-47 set to cardioid, against tone generator at 30cm, front, right, rear, left positions

And here’s an M-Audio Nova, same conditions, for comparison:


nova, cardioid, against tone generator at 30cm, front, right, rear, left positions

The total pickup is certainly omnidirectional. But this is not a true basket head on this thing, there are support brackets on the left and right sides, and that does affect tone! High-end harmonics definitely fall off on the left and right sides.

So while it does qualify as an omnidirectional microphone in “omni” mode, I’d have to call it flawed in that mode, and use something else – like an Oktava 012 with the omnidirectional head, which is built for true omnidirectional pickup and does not exhibit this behaviour.

Alternately, I suppose it might be usable in that mode if you’re recording someone who is a tad screechy – aim the wrong side at the performer and presto, fewer high harmonics. EVERYBODY WINS XD

Mirrored from Crime and the Blog of Evil. Come check out our music at:
Bandcamp (full album streaming) | Videos | iTunes | Amazon | CD Baby

solarbird: (made her from parts)

After a few suggestions that I do so, I finally went ahead and splurged for another kit microphone – the micparts RK-47 capsule and amplifier circuit board, both sized to use the dirt-common MXL-990 microphone case.

The MXL-990 is kind of a go-to microphone for people just starting at home recording. I don’t recommend it, even in that role; I think it’s a nasty piece of kit and sounds terrible. Even at $50 – the used price, roughly – you’re better off spending another $40 and getting a low-end AKG or Nova. But it is about the cheapest condenser microphone with XLR out that you could pretend to treat like a real mic, and that has meant a thriving mod community.

Here are the basic parts included with the kit. I didn’t have the capsule out – the actual thing that vibrates to sound – because it needed to stay in its protective box until installation.

This is technically an MXL-990 mod, one wherein nothing is kept but the case and socket. It’s also a nontrivial assembly, so I’m going to talk about building in this one, and have some sound samples in the next.

You can see the instruction booklet over on the right. It’s quite clear, and has some important commentary, about extra care being needed not to touch certain components any more than you have to, and which parts are particularly sensitive to heat. It does not say up front that about 25 pages in it will suggest that you use a couple of chemical products if you have them, so I ended up having to order some purple loctite and conformance coating.

If I had it to do again, I probably wouldn’t order the conformance coating, because the MDA sheet on that is not attractive. But now that I have it, I might as well use it. If anyone local needs some, let me know.

As you can see, the board is clearly labelled:

For the most part, this is one of those insert-and-solder projects. The tolerances are reasonable, though they were emphatic about working at a meaningfully lower iron temperature than I prefer. If you use 60/40 lead solder, their temperature would be fine.

(I ran at 350°C, which is a little hotter than they recommend, but I use silver solder, and it seemed to work out. In these cases, I solder only one leg from each component, then cycle back to the start, and do the second leg, then cycle around again for the third if necessary. This lets the heat that is building up in the component itself dissipate, reducing peak temperature inside the component. This is most important with semiconductors, usually – particularly transistors.)

They also have you cleaning the board with isopropyl alcohol regularly, and suggest cleaning your iron between each solder point. I don’t know if they just meant wiping the tip or actually using the sal ammoniac, so I kind of compromised. (Though for the most sensitive components, I did do a complete sal ammoniac cleaning between each point.)

This is the board essentially complete:

The process is a bit fiddly, I have to say. They’re quite strict in the manual about flux cleanup and so on, so you’ll end up using tweezers and foam swabs and lint-free cloth and and and. Still, with kits this expensive, it’s best not to take chances.

You might note that the connections on the switch are not made yet; that’s because you’re wiring two of the leads from the pickup to the circuit at these points directly. Those connections are where they’re most persnickety about how things are to be done – but I followed their directions precisely and it worked out.

Here’s the new RK-47 pickup canister in its mount:

I’m liking this canister already, at this point. I know, I know, “good feel” is silly, but this capacitor pickup has a good materials feel to it. The mounting system you see in use is also partly a replacement – the original was a ring mount system; this is a saddle, larger, and is also where you need the threadlocker. I’m not convinced MXL used a threadlocker on that post, but I see how the larger, heavier pickup does kind of call for it.

Once the board is mounted back into the bracket and grille, you end up with this:

And once you’ve wired the three leads from the pickup and the three from the XLR connector, you’re pretty much ready to test.

I did have an initially-worrying test result: voltage at the primary test point – the one you conduct before any audio checks – was running high. The manual had a lot to say about what you’ve probably done wrong if it runs low, and further tests to isolate your mistake, and spends a couple of pages going on making sure you’ve got a bare minimum of 45V going to the exciter, with a preferred minimum of 50v, and an “operating range” of 50-62v, where 60v is optimal.

The range on mine ran from a low of 62.45v to a high of around 68v, so I was concerned. But it turns out the actual operating range is 50v-70v – they just don’t expect anybody to be able to get to the upper end. Support even said that if I was mostly recording quiet material I could safely crank it up to 65v full-time for an even hotter pickup, but, heh, that’s not what I generally do, so I’ll leave it where it is. XD

As for the sticker – I put “MP-SVS” over the “MXL” logo because at this point, it’s not an MXL microphone in any way, so why would I keep that? It’s just silly. MP-SVS is MicParts-Supervillain Studios, naturally. I also changed the model number to RK-47, after the canister number.

I did not mention before that this is a switch-selectable dual-pickup-pattern microphone:

That will let me try an instrument miking style I’ve wanted to try for a while anyway. The only downside of using an existing microphone housing is that the switch has to live on the inside, and you have to unscrew the bottom of the housing to get to it. But it’s just twist-open, so not really an issue.

Regardless, as you can see, the switch says “cart” (for cartioid) and “omni” (for omnidirectional), but I think I’m pretty sure it’s really more figure-eight, given the construction of everything and what the switch does. And pictures in the manual actually agree with that – the printing on the board is different in the photos.

An optional modification to the case involves coating the lower half with silicate caulk (or a couple of other listed materials). I, of course, did that too:

The idea is that this reduces case resonance. Does it? I dunno, I just went ahead and did it because silicon caulk is dirt cheap – the local hardware store down the hill had a small tube for three dollars. The lack of silicon at the top is necessary – the upper part of the case scrapes into the lower half quite tightly, and any added material would be scraped off right away. As you can probably see, it’s even scraping off its own coating.

This is what it looks like all put together, and put in the standard MXL-990 shock mount:

As always, larger versions of the photos are at my Flickr account.

All in all, I’d call it a straightforward well-documented build. It’s far from the most difficult I’ve done. Not a good starter project, but if you’re reasonably good with a soldering iron, you’re going to be fine.

And this is already a long post, so I’ve decided to make it Part 1 of 2, and will talk about the important part – what the result sounds like – next time. Spoiler: a whole hell of a lot better than an MXL-990. But it also has audibly (and visibly, in the waveform) finer and more precise pickup than, say, my Novas.

I’m also going to see if I can’t do some comparisons to my Oktava 319 and possibly the (already-improved via mods) Oktava 012s as well before next time. That, too, might be interesting.

Have a good weekend, everybody!

Mirrored from Crime and the Blog of Evil. Come check out our music at:
Bandcamp (full album streaming) | Videos | iTunes | Amazon | CD Baby

solarbird: (made her from parts)

I thought I was going to be posting about a new microphone I built today! I was pretty sure I could get the kit construction finished and even make a couple of recordings.

And I could’ve, but no, I had to have one of these moments:

…wherein on page 24 of the assembly instructions it talks about connection sealing material and special kinds of thread-locking fluid which aren’t strictly necessary but are definitely good to use, if, you know, you just happen to have them around, and if you don’t, well, you can add them later but EVERYTHING WILL EXPLODE.

Particularly the connection sealing material. Apparently.

None of this was in the components and tools list up front. Of course. So thanks, now I get to order all that stuff, which I already have, and it’ll get here Wednesday. (No, it’s not at the local hardware. I did check!)

Anyway, this is how far I got, and where it’ll be until Wednesday:

Mirrored from Crime and the Blog of Evil. Come check out our music at:
Bandcamp (full album streaming) | Videos | iTunes | Amazon | CD Baby

solarbird: (Lecturing)

All those posts I made about data cleanup and old hard drive data recovery got someone I know from my old SF club interested in his old Amiga hard drives, which he sent me, and which I imaged for him, and did as much file recovery as I could.

One of them was just fine – complete image, files pulled off separately as well, life is good. The other…

This is what’s actually in the disk’s partition table.

And for those who have no idea what that means: disks can be divided up into smaller disks, more or less. There are reasons to do this, mostly on servers, mostly not workstations, but whatever. These are called partitions.

This particular drive is cut up into six, count ’em, six partitions, sort of, except that two are completely broken and one is made up of parts that are also at the same time parts of two other partitions, and just kind of… overlayed atop each other.

Spoiler: this don’t work.

It’s like if you got to chapter two of a book, and midway through, chapter three appeared, overprinted and interleaved with the second half of chapter two. And after chapter two ended, chapter four picked up, printed underneath the rest of chapter three, making chapter four unreadable. But you can still make out chapter three if you squint.

There is no partition utility in the world which would let you do this.

The amazing thing is that I managed to recover the contents of both “TWork” partitions, which is a lot like getting chapters two and three from my analogy back and sorted out separately and readable. Couldn’t do anything for chapter four, though. And chapter six… I don’t know what happened there. It’s some kind of goddamn chainsaw murder scene with horror movie implications.

Sadly, both of these drives did the all-too-common very-old-drives-out-of-storage thing, where I managed to image them and pull contents, but after one last heat up/cool down, they just said “aaaaand we’re done here” ’cause that seems to be that. They’ll work long enough to come online, but the Viking won’t even stay up long enough to format it, and the Quantum ST will format, but will only let you write between 10 and 60 megs before taking itself offline and hiding from the controller until you reboot the machine. So no eBay listings for these.

Still, I’m happy. I made enough auctioning off all this old stuff to buy another microphone kit and I am closing in on another toy. Expect more DIY studio gear posts! It’ll be fun. 😀

Mirrored from Crime and the Blog of Evil. Come check out our music at:
Bandcamp (full album streaming) | Videos | iTunes | Amazon | CD Baby

solarbird: (korra-fruck-out)

This is what’s left of the rubber feet of my 1994 Commodore Amiga 4000/040.

Yep. That’s a viscous tar-like residue that ran along the bottom of the case and into the gap between the bottom case and the cover. It’s very sticky and resistant to detergent, but, fortunately, not to citrus-based sticky tape removers. Isn’t it gross? I mean, seriously, doesn’t this look like some sort of OOZE OF SATAN special effect?

Oddly, the stickytape used to hold the rubber feet on was intact! And, as far as I can tell, fine. This actually saved me a lot of cleanup, because the cores of the old feet were pretty solid? The further I removed down, the more like rubber it became. The thickest material had an actual rubber-like consistency.

Fortunately, all the rubber that ran could be cleaned up pretty easily with an old spudger and a lot of orange oil. But goddamn.

I’ve repaired tube equipment from the second world war and this is the most bizarre materials degradation I’ve ever encountered.

Mirrored from Crime and the Blog of Evil. Come check out our music at:
Bandcamp (full album streaming) | Videos | iTunes | Amazon | CD Baby

solarbird: (molly-oooooh)

I wanted to post about the cool second-generation crystal microphone today BUT NO IT’S ALL STUPID AND NOISY AND I DON’T KNOW WHY but it sounds like either a really bad cold solder joint (please be that) or a bad transistor (@&*$&!!! special orders please don’t be that) and I don’t know which.

It’s too bad because I came up with a nice little jury-rig jig (say that five times fast) and so the backplate of the housing came out really well and I was looking forward to showing that off. Fingers crossed this is some sort of Surprise It’s Easy! fix – that would indeed be a surprise, to be honest about it, but a pleasant one.

In the meantime, enjoy this video of Overwatch players in custom game mode making some genuinely gorgeous Genji Beams. These are effects created by lining up opposing teams of Genji players opposite each other, in continuous-shot-deflection mode, and hitting them with various weapons. The shots bounce back and forth between the teams, and you get some really neat graphics interactions. It’s pretty cool and occasionally hilarious. Enjoy:

Mirrored from Crime and the Blog of Evil. Come check out our music at:
Bandcamp (full album streaming) | Videos | iTunes | Amazon | CD Baby

solarbird: (made her from parts)

I’ll be at Orycon this weekend! I have panels and workshops and a concert on Sunday afternoon, so please come out for it!

They’ve actually given me a panel on building old-technology microphones, which is pretty cool, and I’ll let people record themselves on a 1920s-ish carbon mic and a 1940s-ish crystal mic, so hopefully that will be fun. People really seemed to enjoy it when I had the setup running at my dealer table at CBCC.

I need to find time – somehow – to wedge in building a new amp board for the microphone panel, because I don’t want to take apart the existing crystal microphone for show-and-tell, and I think it’ll be neat to show that off. Later today. Hopefully. Assuming nothing else explodes.

Like, Monday, right? Things Happened, and then I got to spend the afternoon pulling unimaginably gross material out of a fluid pump system while going, “oh gods, oh gods, I hope this is only cat litter, please let this just be cat litter,” and when that’s happening, you have officially reached what one might call a BAD SCENE.

Not “darkest timeline” bad scene, but: bad.

so disgusting

Anyway. Thank the gods that’s over. See you this weekend!

Mirrored from Crime and the Blog of Evil. Come check out our music at:
Bandcamp (full album streaming) | Videos | iTunes | Amazon | CD Baby

solarbird: (music)

Oktava has some great microphone designs. But the quality of the components can be pretty random, particularly in the used market, since a nontrival number of those were made in the early post-Soviet era. My two 012s sounded pretty different – one in particular rather unpleasantly harsh – so I implemented Recording Magazine’s recommended component upgrade* on the harsh one, which we’ll call Nr. 1.

Nr. 1 may have been modded a bit before. It’s certainly been opened before; one of the three screws was stripped and useless, the other was jammed badly. I had to drill both out, so I’m hoping I can order replacements. The third was fine.

As soon as I had the microphone open, I saw what Recording meant by random components. The key transistor was a make so old that it had a metal shield ground cap, and separate lead to that cap, something I haven’t seen in gear made after about, I don’t even know, 1978? I also saw what they meant by “fragile circuit board,” because wow, you could lift these circuit board traces with an overly-aggressive hair dryer. Take care!


Comrade!

Still, it was mostly a matter of being methodical and not rushing things, and in good time, I had the key components upgraded, with no surprises other than the transistor’s extra lead.

These are three unmodified before/after snippets in one recording – recorded under identical conditions other than the internal microphone electronics – of Oktava mk-012/mc-012 nr. 1 in my studio. Even on laptop speakers, I can hear the harshness, particularly in the first sample. In all cases, it’s pre-modification first, then post-modification after:

Oktava MC-012 nr. 2 sounded very different to nr. 1, before; opening it, I could see that the components used were of a significantly more modern variety. It may well have been made later, which would be part of that. Now, the two microphones sound much more like each other, indicating that nr. 1 really was meaningfully different in component quality.

Here is a recorded comparison of nr. 2 (still factory) and nr. 1 (upgraded). These recordings were made simultaneously, with the two microphones right next to each other. The differences are much subtler, but I think the upgraded nr. 1 has a bit more presence – or maybe sense of stage – than the factory nr. 2. Despite being mono recordings, it’s almost like there’s a slightly better stereo image in the modified nr. 1… but give a listen and hear for yourself, see what you think.

You’ll definitely need headphones to have any chance of hearing anything interesting here. Factory nr. 2 comes first in all cases:

So, all in all, very glad I did this to nr. 1; pretty sure I’m going to go ahead and do it with nr. 2 as well, though I expect a much less dramatic change.

The only thing I’m thinking about now is – there’s a bank of capacitors in back. They’re good ones – Philips, not generic, which have a good durability and spec-compliance record. (I don’t know whether they’re original; some Oktava 012s shipped with quality caps already in place, and their track record has improved with time.) So I shouldn’t need to upgrade them – and the article at Recording Magazine says not to bother if you already have “improved” capacitors.

But I don’t know how old these are, and electrolytics have a lifespan. That’s measured both in calendar time (years), and in use – tho’ the latter is in tens of thousands of hours, and these mics are certainly nowhere near that.

The small downside is time spent, the large downside is the possibility of circuit board damage, which wow I don’t want. The upsides would be 1. possible sound improvement if they are aging already, and 2. Never having to think about it again, in practical terms.

So I dunno. Get it out of the way, or leave sleeping caps lie? Hm.
 
 


*: errata for the linked article: Capacitor “C6” in the parts list is actually capacitor C1; there is no “C6” in the build description or circuit diagramme; I assume this is a typo.

Also, some of the items in their parts list are no longer made, but they have exact replacements. R1/R2 exact replacement part number as per my October 2016 Digi-Key invoice: MOX200J-1000ME-ND. Capacitor C1 (listed as “C6” in parts list, see previous paragraph) current part number: 445-4737-ND. Capacitor C2 current part number: 399-1418-ND. Capacitors C3 and C4: 4073PHCT-ND. Capacitor C5: 4047PHCT-ND. Mostly, the substitutions are lead-free versions replacing earlier versions with lead.

Mirrored from Crime and the Blog of Evil. Come check out our music at:
Bandcamp (full album streaming) | Videos | iTunes | Amazon | CD Baby

solarbird: (made her from parts)

I made a thing! It’s called a focus knob. It’s quite simple and normally you’d built it into an electric guitar as a guitar mod, but since I don’t have an electric guitar, I built it as a pluggable external box.

Basically a mild high-pass filter that serves to pull out ‘boominess’ from instruments, it puts a bit more of an edge on an instrument’s sound – the more you turn it up, the greater the change. As effects go – on my zouk, anyway – it’s pretty subtle. But it’s also the kind of shift that is multiplied by later effects added in, and changes how later-in-chain boxes like distortion pedals work.

As you can see from the instructions here, the wiring takes all of about 20 minutes’ time. But it’s good warmup for making a bunch of component upgrades to my Oktava 012s, and I already had all the parts from the Great Radio Shack Lootfest of 2014. Plus, hey, cool hard candy tin!

So I have a a HARD CANDY knob now. It goes from Hard to Candy. Yum. 😀

Mirrored from Crime and the Blog of Evil. Come check out our music at:
Bandcamp (full album streaming) | Videos | iTunes | Amazon | CD Baby

solarbird: (molly-oooooh)

Several months ago, I saw online a prototype of a sampling synth with waveform editing and a uniquely cool physical user interface. I don’t remember whether I blogged about it at the time or not, but I certainly talked about it on social media and such.

It’s not so much that it does anything you can’t already do; you can do everything it’s doing with a modern digital audio workstation, for example. But the physicality of the interface looked delightful, and that sort of thing really, really matters in instruments. Including synths. It made sample synths look fun to play in a new dimension – one far more instrument-like than I’d seen before.


Collidoscpe v2

I wondered at the time if they were looking for a commercial application, to build them to sell. But if they were, they’ve ditched that: it’s gone open-source. Not just source code for the software, but instructions and 3D CAD files if you want to build one yourself to their physical specs.

Admittedly, the case-build instructions are… a tad sparse. But that’s half the fun, right? Component-wise, it’s basically a cakewalk. (Silly me thought the waveform display was some fancy custom thing HAHAHAHAHAHA NO IT’S A STANDARD LCD MONITOR BEHIND A FRAME HAHAHAHAHA etc. But that’s the smart way, so.)

Anyway, yeah! Project!

h/t: Klopfenpop for the link

Mirrored from Crime and the Blog of Evil. Come check out our music at:
Bandcamp (full album streaming) | Videos | iTunes | Amazon | CD Baby

solarbird: (Lecturing)

Well, that was neat – the “warmup” storm was the big one, the big one swerved north at the last minute and weakened, mostly missing us, and then surprise-collapsing over Victoria. Still, we did lose power and were offline most of a day, so if you missed Friday’s post about recovering damaged recordings – a bit of a geek-out, really – that’s what happened.

I took advantage of the unexpected uptime to finish up that project, by the way. It was interesting, and I learned stuff, like usual. The condition of the tape (and damage in the recording) varied all over the place, and arguably too much even to try to split it up into a million shards. Though there was a lot of this kind of crap:


This is one phrase of lyrics

That’s a combined automation of level and compression ratio. Here’s what it looked like over a larger area, about 3/4 of the way through. Yes, meaningfully more got added to this:

And the sad thing is, that’s just me trying to attain listenability throughout. I’m not trying for “good” – that isn’t attainable, but less noise and less distortion and fixing dynamics over time, that I can do.

Some of the tape wasn’t really that bad! I mean, it’s a 23-year-old cassette recording made on some sort of portable device set on what sounds like two different autolevelling schemes – it changes once when the recording was stopped briefly, I suspect the operator changed modes and I really wish they hadn’t – so “not really that bad” comes with a lot of caveats. But still, not that bad. Lots of hiss, lots of tape rumble, small dropouts, and so on, but not unintelligible. Fatiguing to listen to over time.

Here’s a short sample of “The Crawl,” early on, direct from the tape. Hissy, some sort of mid-band distortion that isn’t too bad in short doses (but really gets annoying over time), off-centre sound placement. But otherwise not that bad. You can hear stuff.

So I ripped the hiss off, did some work to improve dynamics, pulled out what I could of the distortion, threw on EQ to bring back out the low end, re-centred and smoothed it a bit, and here y’are.

Then there are other sections. After that mode switch got thrown, the whole recording got weirder. “The Profiteers” was particularly bad. Here’s the original. I know the lyrics and I still can’t make them all out here. But I can in the restoration. It’s not good, but you can make things out. This is where I needed that whole stack of plugins I talked about on Friday.

And just as importantly – and something short samples like the above won’t give you – it’s listenable over time. Some of these problems are really hard to tolerate over the two hours of this recording. They’re not bad in short doses, but they grate.

Like, the original seems to have more high end, right? But it’s not real. Eventually your brain figures out it’s just hiss, with your audio centre filling shapes into it, and it’s wearying. In short comparisons, the brightness is attractive, and the restored version sounds kind of dull in contrast at first – but as with light, your ears adjust to recordings, and that goes away with listening.

Similar are all the damn-autolevel-to-hell level changes. They’re not necessarily so bad in short doses – some of those are like punches to the face, but most aren’t. But even with that, EVERYTHING REALLY LOUD PUNCTUATED BY surprise underlevelling is also wearying.

So the restoration maybe doesn’t sound good, on any kind of normal scale – but I got it to the point where, particularly on laptop speakers, it sounds pretty okay. I can listen to it. Occasionally – just occasionally – it even sounds musical. And there’s enough there there to remind me how much I miss this Great Big Sea.

There was one thing I couldn’t fix though, no matter how I tried. And that’s during “Excursion Around the Bay,” wherein early in the song, some fucker orders espresso at a George Street bar. And so you get that espresso machine foam blast noise right in the middle of a verse.

WHAT. THE. FUCK. YOU WANT COFFEE, GO TO STARBUCKS.

Gods damn you, espresso man – gods damn you.

Mirrored from Crime and the Blog of Evil. Come check out our music at:
Bandcamp (full album streaming) | Videos | iTunes | Amazon | CD Baby

solarbird: (molly-oooooh)

In yesterday’s post, I posed a question: do USB chipsets matter in the 2.0 environment? I had reason to suspect they might.

The answer is holy crap yes they matter they matter so much it is unbelievable.

First, let me talk about what prompted this research, so you’ll know why this matters.

On my old sound interfaces I had live monitoring in hardware, so I didn’t have a lot of need to care about latency. Since that won’t mean much to most people, I’ll explain; when recording, it’s good if you can hear yourself, in headphones. If you’re multitracking, it’s critical.

My old audio interfaces did this with direct connections in the hardware. Whatever came in the microphones also went out the headset. There are advantages to this method, but also disadvantages, in that you aren’t actually hearing what’s being recorded, just what’s being sent in the microphone jack.

But now, I have this shiny new 1818vsl, which doesn’t do hardware monitoring under Linux. Higher-level kit generally doesn’t provide that; they’re assuming you have enough computer that your computer can send back what is actually being recorded, effects and all, and that you’ll do that instead.

This means I now have to care about latency in my system. Latency is basically delay, between mic and computer, and computer and headset. And if the computer is feeding my monitor headphones, that delay matters. You want to hear yourself live, or close to it, not with, oh, a quarter second of delay or something horrible like that.

Now, the good news was that straight out of the box on Ubuntu 16.04 (the latest long-term support version), I had better, lower latency numbers on my new 1818vsl than on my old hardware, when I was using that on 12.04. I could get down to a buffer size of 256 samples, and three frames, which gave me about 30ms basic latency – roughly half what I had with my old hardware and old install. I could use it as-was.

But I couldn’t go any lower on those buffers. One more setting down, and even playback would lag. It’d be okay until the system had to do anything else, then you’d get a playback pause, or a skip, or if recording – presumably, I didn’t bother trying – lost sound. That’s unacceptable, so 30ms was the lower limit, and I wasn’t sure it was a safe lower limit.

And that’s what got me doing all that chipset research I talked about yesterday, and I ordered a new USB card (plugs into PCI sockets) based on that research. I was hoping for a couple fewer milliseconds of latency, that I wouldn’t actually even use; I just wanted a safety margin.

So that new card arrived on Sunday, with its OHCI-compliant chipset made by NEC, and I popped it into the machine and started things up with normal settings.

At first, I was disappointed, because I only saw about half a millisecond less lag, instead of the 1-2ms drop I’d hoped to see. But across tests, it was more consistent – it was always at that same number, which meant I could rely on that 30ms latency in ways I wasn’t sure I could before.

They I decided to see what would happen moving the sample buffer setting one level lower, into what had been failure mode. And the result was 1) it actually worked just fine, where it hadn’t before, and 2) when running analysis, tests showed much lower latency at that setting than with the previous USB ports.

That was an ‘oh ho‘ moment, because it implied that the 256-sample run rate was basically the spot at which the on-motherboard USB could just keep up, and trying to run faster wouldn’t actually produce any actual processing improvement. It’d try, but fail, and time out.

So I did a couple of recordings on that, and they all worked. Then I dropped it another level, until finally, I just said hell with it, let’s just set it as far down as the software will allow and see how hilariously we explode.

I just successfully recorded test tracks four times with these settings, on the new card:

0.7 milliseconds isn’t even something you think about on USB 2.0. 2.8ms, maybe, okay. I’ve seen that managed a few times before, and that’s genuinely indistinguishable from realtime/hardware monitoring. But 0.7ms?

Seriously, this is well into “…is that actually possible?” territory. I’ve never even heard of someone running over USB 2.0 at latencies this low.

So, I guess it looks like the chipset matters a whole lot. Maybe not for most applications, and maybe not in the same way as in USB 3.0 or in FireWire, were there are serious compatibility issues. But in the 2.0 world, in realtime audio, it appears that the chipset makes all the difference in the world.

And yet, I can find this nowhere online. I’m beginning to think nobody bothered until now. Certainly when I’ve asked about it, the response has “why are you on USB get firewire” or “why are you on USB get PCI” because sure I want to throw out all this hardware and start over THANKS NO.

I think USB users have been trained just to accept it and deal. But surprise! You don’t have to! You can actually get a better USB card, if your system allows it, and it’s $30 instead of $1300!

So, HELLO, OTHER SMALL-STUDIO MUSICIANS! You want a chipset that uses OHCI on the USB 1.1 level even if it’s a USB 2.0 card or later because the 1.1 layer still matters, and still gets invoked by the higher-order drivers for card management. See previous post for why that’s important.

This means avoid Intel and VIA chipsets, and look for NEC or SiS – or anything else that loads OHCI drivers and not UHCI. If you’re on Linux, you want to:

cat /proc/interrupts | grep usb

If you see “uhci_hcd” in there, you have a UHCI chipset running your USB port and getting a new USB card with an OHCI-compatible chipset (and disabling whatever’s already installed) might help you with your latency issues.

Good luck!

Mirrored from Crime and the Blog of Evil. Come check out our music at:
Bandcamp (full album streaming) | Videos | iTunes | Amazon | CD Baby

solarbird: (pingsearch)

I’ve been trying to find out whether there’s any sort of difference between USB 2.0 cards, specifically as addresses the needs of digital audio workstations on Linux.

Very few people in linux communities seem to have addressed this question at all, and none I can find on the audio side. (Firewire, oh my gods yes – huge lists. Just not USB.)

But I did a lot (a lot) of digging, and discovered via the Linux USB kernel driver dev mailing list(!) that while there’s not much difference on the USB 2.0 side, there are important differences on the 1.1 side. These difference manifest in two different driver models. That still matters at least a little bit in 2.0, because those 1.1 drivers still get loaded.

Anyway, that difference is that there are two very different driver interface models. One is UHCI, created by Intel and used by Intel and Via, mostly. The other is OHCI, which Compaq pushed when it was still around, and Microsoft preferred; it has less intellectual-property load, and NEC, SiS, and some other makers use it. If you see a “Mac compatible” card? It’s going to be OHCI.

The OHCI model puts a lot more of the business of doing USB into hardware on the card. UHCI has the processor do that work. And while that isn’t a heavy load, it is a nonzero load, and more importantly means that UHCI chipsets require more CPU attention than OHCI chipsets, on a recurring basis. And that is something we don’t need in a digital audio workstation; there are only so many board interrupt opportunities; I want them for moving data, not servicing USB mechanics.

Once I knew that, I did more searching and found people saying how switching to a NEC chipset card had (in one case in particular) ‘saved their bacon’ specifically on their digital audio workstation. They were using ProTools on Windows, not Linux, but it was still with a USB audio interface.

The chipset used by my on-motherboard USB ports is, of course, Intel, and therefore UHCI. (And UHCI drivers are actually loaded, I checked.) There’s also an on-motherboard hub between the outside world and the one true root device; that doesn’t help anything either. So there’s a nonzero chance I’ll see improvement both from changing from UHCI to OHCI, and from moving to a true root USB device instead of a hub device. It won’t be much, but I’m only looking for a few milliseconds of latency here. And even that’s more for… reliability buffer, I suppose? Yeah. Reliability buffer, rather than pure necessity.

I’m mostly posting this 1) so I remember it and 2) so other people looking for this data can find it. HI! I can’t be the only one!

I’ll update this post if I get interesting results.

eta: INTERESTING RESULTS AHOY: CHIPSETS MATTER SO MUCH OMG. I’ll write up a post with details, post it tomorrow.

Mirrored from Crime and the Blog of Evil. Come check out our music at:
Bandcamp (full album streaming) | Videos | iTunes | Amazon | CD Baby

July 2017

S M T W T F S
      1
2 3456 7 8
9101112 13 14 15
16 171819 2021 22
23 24 25 26272829
3031     

Most Popular Tags

Syndicate

RSS Atom