Saturday, October 25, 2014

Somewhere...

Somewhere out there is a theater where the Lighting Designer gets no support. Where they are asked to rig the lights, repair the inventory, program the cues, run the light board without a single assistant. Where they are "accidentally" left out of meetings and schedule updates. Where they are expected to draft a plot without ever seeing a ground plan, focus without seeing a set, write cues without actors or blocking. Where they are given limited access, and when they ask if someone could please turn out the work lights so they could see what they are doing, they get laughed at. Where the inventory sucks so bad they have to bring in their own equipment as a free loan, and then watch helplessly as set movers smash into it, set painters splash it with paint, and helpful stage managers randomly pick up chunks of it and stuff them away in closets without telling anyone.

At this theater, incredulous laughter greets the idea that it actually takes more than a couple minutes to focus a plot, and it is a job that requires more skill than the random volunteer doing their community service. And when the Lighting Designer bothers to speak (having somehow not digested the way their emails go unanswered, their questions ignored, their concerns brushed off) every random person speaks up with what to them is the simple obvious solution. The Lighting Designer is trying to fill the stage with moonlight? "Why, just stick a blue gel on a 6" fresnel somewhere and voila!" they say, "I really don't understand why you have to make everything so hard." Except, of course, when they hit Opening Night, and the lighting sucks -- oh, then everyone blames the Lighting Designer.

I'm sure such a theater exists for Lighting Designers. There are some really stupid theaters out there. Thing of it is.....ALL theaters are like this for Sound.

Friday, October 24, 2014

For Archaeologists Who Have Considered Suicide When the Scion is Enough

And just to round things off, started Tomb Raider: Anniversary. That's the Crystal Dynamics (the first reboot group) remake of the very first Tomb Raider game.

It is definitely the hardest of all the Crystal Dynamics series. The controls are not even quite as fluid as Tomb Raider : Legend, and the camera screw is much worse. Particularly awkward are the wall runs; in those, the direction Lara will leap depends on the exact instant you jump, as the camera changes axis at the apogee of each swing.

The first two games give insight to just how mature Tomb Raider: Underworld was; much more fluid animation, better combat mechanics, minimal camera screw. Although Legend includes several vehicle sections, only in Underworld is the vehicle integrated into the same environment as Lara, meaning you can drive over the same terrain you can run over. And also meaning you can use the vehicle in melee.

Underworld also did the best job at integrating story and tomb exploring. The puzzles were, unfortunately, simplistic, but the tombs were monumental and the isolation palpable. For all that the tombs of Anniversary are equally without human presence (aside from your own and the occasional appearance of a story element), there is a distinct lack of context, of any scenes outside the tombs that can really give you a feeling of having pushed far away from the world outside. Instead it becomes a series of puzzles in some interior space, as hermetic as the (rather more intentionally so) Portal.

The puzzles in Anniversary are difficult and few of them are the contrived "shoot the beam that against all logic falls across the stone to knock it into the lever" types. There's been only one so far that I stopped and went for help on, though. And, pity -- I was doing exactly the right thing already, but the quirky controls were making my action fail.

The oddest reality break in Anniversary occurs in large part because of the hub structure of many of the puzzles. Frequently, a mistake will send you back to a common place where it will take you a long and frustrating time of recreating your previous moves. Or...you can step off a platform or into a whirling blade, and restore to a more convenient save point.

Literally, suicide is painless (compared with the alternative).

I think personally the hub nature of the puzzles doesn't help. In Underworld, you always had the sense of moving closer to your ultimate goal. Even in some of the more hub-based systems, you were visibly progressing across a long hall or up a huge structure. In Anniversary, you end up crossing a room multiple times; going up to collect a key, back down to open a door. This makes it feel less like you are accomplishing anything, and makes you impatient to finish that room and move on.

A number of the puzzles involve a similar lack of progression. More than one "flooded room" appears that requires you to climb down the room progressively turning taps to lower the water, do something at the bottom of the room, then climb all the way back again turning the taps back on again. And repeat this several times if you didn't position the thing at the bottom exactly right for where it needs to be when you are finally back at the top.

They even nest; one door might take three keys to open, which are each held in a hub of their own that takes multiple keys to open, each of which is, of course, some complicated little problem of ramps and spinning blades and so forth.

When you come down to it, it is boiled down to mostly pure gameplay. But considering that the story elements, character interaction, and so forth in the other games is extremely scripted and minimally interactive -- little better than watching a cutscene -- I'm not entirely sure this is a bad thing.

A Feeling of Power

Bought a couple of those stick-on LED dome lights for the rather dark Orchestra Pit. Turns out they were designed to turn off automatically after five minutes. So I opened them up, and after figuring out the circuitry, cut the traces to the chip and soldered a jumper. Now they stay on as long as you want.

There's a real feeling of power in doing that. A lot of tech is going the Apple direction these days -- in which it decides it knows your application better than you do, and will actively get in the way of you using it the way you actually need.

Also repaired a wireless microphone I'd bricked myself. The mini locking connector for the microphone element is under a lot of stress in use, and tends to snap inside. I've learned the way to remove the old connector is to dike it into little pieces while still in place; then you can carefully de-solder the remaining bits and solder on a replacement jack.

The first time, though, I tried to pry the whole thing off, and ended up tearing the traces right off the PCB. Well, fortunately, all of the traces to the jack pads also lead to test points. And the test points are just big enough to take a drop of solder and the end of a jumper wire:



That's the tip of a miniature screwdriver visible there in the blur. This thing is small, and tightly packed. But the repairs passed the bench test. Next is to hook it up to the sound system and see if it is quiet and resists vibration.


Oops.  So it did work after all -- after re-seating the RF daughterboard. But there's a shot button on the motherboard as well and I'm not quite up for reworking an SMD button. 



And around again. Turns out you can repair one of these SMD buttons in situ; I carved a hole in the plastic with an X-acto, slipped the tip of the blade under the metal, and pushed up to crimp it slightly. Works now.

However...turns out the board has low RF. Maybe something shorting the antenna?

Thursday, October 23, 2014

Poppins_02: Development

In a show that went through a proper development process, I would have had a concept meeting with the Director, firmed up some ideas, and consulted with the Music Director about whether some of the cues are pit-generated or pre-recorded effects.

The next step would be a proper spotting session. I've done this on my own, and with the Director. The best and simplest is just to talk through the entire show with scripts open, and mark each moment with a bit of colored tape. I'd insert a picture here ...but my Poppins script is at the theater.

As a lighting designer, I'd underline the actual line or action in the script, with a line pointing to a circle within which appears the number of the light cue. Sound is more fluid; unless I'm charting for someone else, I just mark the approximate line. Often as not, these days, I don't even use cue letters. I just name the cue in QLab.

After trying a few options I've settled on colored plastic tabs from Post-it. They're pricey but I get a few re-uses. I follow an arbitrary scheme where one color is background ambiences and scene shifts, another color is spot cues, and a third color marks vocal processing and live improvisations.

With this in hand, you sit down at the Paper Tech. This is where Sound, Lights, and Stage Manager sit down to hammer out their cue numbers, with the Director there to keep the concept of the show's flow intact. Or so goes the theory. Practically speaking, a musical is such a huge lighting beast (250+ cues is typical) the entire 4-6 hour meeting is taken up with the Lighting Designer talking as fast as they can while the Stage Manager scribbles the data down in their script.

If sound is called cues (aka, if they are taken by an operator on headset) or if the Stage Manager is taking the cues themselves, then you need your cue list to go to them during this meeting.




There's been a bit of a pendulum going on with this of late. When I started in theater, large crews were the norm. Stage Manager would call. Light Operator and Sound Operator would run the boards. Then budgets shrunk, and the standard began to put the Stage Manager in charge of running light cues or sound cues. Sometimes both!

But then, lighting grew steadily more complicated. The standard for musicals these days is for a new look on almost every measure of music. And the smaller theaters are adding more rails and other moving scenery -- which also have to be called by the Stage Manager. And video, not infrequently; yet another set of buttons to press.

So sound is moving away from being one of the many tasks of the Stage Manager. Yet, with the shrunken budgets still in effect, this means sound cues get dumped on the plate of the already overtasked FOH mixer.

In my own design evolution, I think of many parts of the soundscape in increasingly musical terms. For Poppins, in fact, a majority of the sound cues are not going to be cues at all. They are going to be a performance.




Again, when a show is going smoothly, you can program in rough cuts of cues at this point. In some shows there may be a dry tech (set movements, lighting, and effects all without actors) but the usual is to go direct to Cue-to-Cue. This is where every part of the show is done all together with the exception of most of the dialog and singing. Or most of the costumes.

Cue-to-Cue is called this because that's the literal process; you go from a few lines before a scene change or the start of a song or a flight sequence, and keep going for a line or two until the Stage Manager yells "Hold!"

For all practical purposes, you don't do much in sound for this. There's a lot of lighting, and a lot of set movement, and of course all those annoyingly fiddly practical effects, to get through. I tend to blow off most of my cues except for those that are needed by the actors, and spend the time building sound effects. The most important thing I do all day is make sure the Stage Manager has a god mic.




But that is jumping ahead.  Speaking effects, some shows really are bread and butter. Phones ring, OS toilets flush, there's a train station announcement. But I've discovered the majority of shows seem to have a distinct flavor, and that is often systemic in how it changes the effects and frames the entire conception.

For "A Little Princess," all the "Doll Magic" -- basically, anytime something mystical and unexplained happened, from the food appearing to the Timbutoo dream sequence -- was done with African Percussion. There was a lot of that palette in the orchestra anyhow, but specific effects moments I did with pre-recorded sounds (largely marimba and rattle).

For "Mulan" I executed almost all of the sound effects as if performed by a Beijing Opera orchestra; with gongs and bells and wood clappers. Some of the effects were sweetened a little or used more sampled material -- a little wind noise, a little fire crackling, the scrape of a sword.

In two very different directions, "Starmites" was largely synthesizer effects, and many of the spot cues were manually triggered from a MIDI keyboard instead of being played off QLab in strict sequence. "Charlie Brown," on the other hand, was standard spot cues -- created, however, largely from recorded children's voices.

"Poppins" has been incredibly hard to conceptualize. We've been going around for three or four weeks on the concept of wind. BERT keeps talking about the wind as he Greek Choruses his way around the outskirts of the plot, and MARY of course enters at least once with wind.

My problems with the idea are two; that the orchestration is so dense and so continuous, and yet really wants open textures and silences; and that I'm getting really tired of wind cues (I had wind playing through entire songs of "A Little Princess.")

Well, we did the first Set Rehearsal last night. This is usually done with running crew, and on larger shows is scheduled before Cue to Cue because figuring out how to move all that scenery takes up many precious hours. Idiot theaters will try to figure this out during Cue to Cue, and the really bone-dead stupid ones will try to run changes for the first time in black-out conditions, screaming "Go faster! Go faster!" all the while. Fortunately this company knows better than that!

In any case, the whirling scenery as it moved (relatively!) fluidly from location to location finally sold me. Sold me big time. I haven't even mentioned it to the Director yet but I know she's going to support me here.

I'm going to perform the wind. Rather than try to time out the set changes (which will change through the run) and create pre-recorded cues to suit, I'm going to set up a variety of hand-rolled wind patches on my keyboard and improvise, live, whatever the wind seems to want to do from night to night.

As part of the hybrid nature of this show, it is very nearly through-composed (the music practically never stops) and many of the magical "moments" are specifically underscored. I have a few crash boxes and thunderclaps, and I'm going to do something in environmental sounds to set up the Park and so forth, but these are all in the nature of sweetening.

In "Tarzan" CLAYTON is supposed to fire on a specific beat in the music. In "MULAN" the final fight scene is also set to four specific moments in the music. This show, pretty much everything except ROBERTSON AYE's kitchen accident happens in the score, on a specific musical beat.

Which makes it natural that the bulk of the "effect" for MARY doing her magic is something done by the keyboard player. My only part in these will be to add a little general sound effect to sell what it is her magic accomplishes, whether awakening a statue or rising up a chimney.

And that means, even more so than the usual musical, I can't write any cues yet.

I can not write sound cues without hearing the orchestra play -- any more than you could write light cues without seeing the painted set, or design costumes without knowing what the cast looks like.




Since sound is a multi-hatted job, I'll follow this post up quickly with one about the development of the wireless microphone plot, and the issues with orchestral support and monitors.


Wednesday, October 22, 2014

Makey Makey

East Bay Mini Maker Faire has a lot of amplified sound. The music stage is handled by a local sound company for a sweetheart deal, and I don't envy them -- there's a challenging bunch of groups to set up for. The smaller demonstrations and makers and so forth (like Game of Drones) bring their own PA (the school also has a few old speakers lying around). Between these extremes is me.

I'm covering 2-3 lecture rooms and presentations with whatever I can beg or borrow or cobble up. Year before last was insane; we were spread out all over and weren't given access until the day of the Faire. So I had to hand-carry self-powered speakers and other heavy audio gear across a crowded Maker Faire to one of the distant classrooms where I was setting up. I literally plugs in the mixer, turned on the speakers, popped batteries into a lapel mic and handed it to Chris Anderson as he went on as the first presenter.

Fortunately I'd guessed right in enough of my connections and settings, I had sound.

Helps a lot to know your gear, and know the acoustics. I've dialed in a rough mix on a band without actually turning on the power. I like to think it looks kind of cool to outsiders, when you walk up, select a mic and place it, walk back to the board and get a decent sound. (But then, outsiders really don't understand how incredibly critical having the right mic in the right place is. As I've said before; the right mic in the right place and a mix is all but done. Wrong mic in the wrong place, mix is all but done for.)

I went simpler this year and set up a mono system in the big room, with the mixer nearby instead of trying to set up an FOH-like position for it, and hooked in a pair of handheld wireless with a lav as backup, and a mono connection for people to plug in laptops for movies or slide shows with sound.

The outdoors stage is always problematic because the presenters there don't have outside voices. So we give them a lav and a desk mic and crank up the gain as far as we can. I didn't even bother with two speakers this time (the audience spread was narrower than in previous years and one speaker covered them adequately).

Then an hour into the show, a request came down to help out a children's string quartet. Since I was being simple, all I had was the extras I'd thrown in the box.

Single large-diaphragm condenser on the one tripod boom stand I'd brought. Passed through a Behringer micro-mixer only because of needing the phantom power. Put the mixer on top of the cardboard box it came in and set that in the dirt. Stuck the first speaker they asked for -- JBL Eon on a speaker stand, aimed towards the causeway to entice people over to listen to the quartet.

But since the audience was mostly congregating on the path, I dashed back to where I had the gear stashed, and made one more trip with a Jolly5 miniature (but powerful) powered monitor, and set that up on a chair directly in front of the quartet.

When I set out the condenser, I had it on omni to reach the whole quartet. I was also getting a lot of chatter from passers-by and other extraneous noise (including feedback). So flipped it to figure-eight, and turned it so one lobe was pointed at the talent and the other pointed at the sky.

And it was a decent sound. Pulled in all four strings and not too much environmental noise.

Friday, October 17, 2014

That "Aha" Moment

Got up, jumped into the shower, and while still waking up mused a little on Tomb Raider, and on the (stalled) fanfic. And boom -- realized if I put the SG1 usual suspects in a meetings, Doctor Frasier can mention a connection between what she's been working on for several chapters now, and something she found in Lara's bloodwork. Presto; reason for them to talk to her, and not so much information I give away the whole plot. And as long as they are all there, would be hilarious if Airman Harriman (the "Chevron Five encoded!" guy) is a Lara Croft fan. So I can open the meeting with him dropping a handful of magazines and books on the table.

Thought about Hammond's reaction to her hide-out weapon (gave her a North American Arms "Black Widow" while she was clambering about above Colorado Springs) and that led me to Carter saying she'd go armed as well on the "rather sketchy" trails up there (the words of at least one local resident, according to my research). Which led me to wondering why Miranda would jump into a stranger's car, and suddenly I'm in the middle of my stalled novel instead. And the obvious joke someone can make around my male protagonist that this isn't an urban fantasy, and he's not a werewolf. Besides, Miranda isn't fleeing a clan of vampires; she's fleeing elves. So what would that make him?

A dwarf, obviously, says this someone (probably the as-yet un-named former girlfriend, research biochemist, filk singer and general snarker who is becoming a more and more core character the further I go). Which hooks in perfectly with the various flavors of tech-v-art dichotomy that are going on here. And sets up the third act reveal -- Duergar (which is the band name, but is also the real antagonists; elves who are a little more comfortable with technology, removing my male protag's home ground advantage just as he realizes what is really at stake.)

Oh, right. I should really explain what the heck the novel is about before I go yammering about it. Maybe some day. I just had to write this down because it helps me think more clearly when I do. Seeing the words on the page, I can see better where they work -- and where they don't and need to be revised.

And maybe, with a lot of luck, today will see that flash of inspiration that pulls together my design for Poppins as well.

Thursday, October 16, 2014

Dynamiting Fish

Was just helping someone at the RPF rig up a Cree LED to "pulse" for a prop they'd built. Used an Arduino, because we had one.

Which is like dynamiting fish in a bucket. It is a deceptively simple problem; detect a trigger button, turn on an LED then dim it out. If it were an incandescent bulb, a capacitor would do it all. But LEDs can't really be dimmed that way. You need to PWM them; that is, you switch them on and off too rapidly to see, and you change the ratio between the time it is on and the time it is off.

Which you could do with the old 555 -- possibly the most-used IC ever after the 741 opamp, and certainly one of the most elegant chips ever. Use an external capacitor to change the timing as it discharged. I used to do that trick when I had 555's as sound generators. Even hooked one up to a resistor ladder attached to a binary counter. Made a wonderfully complex sound.

But you'd have to hold the button in until the LED turned "off" completely. So use the 556; the dual timer in a 14 pin DIP. One is operating in one-shot mode and detects the button and then holds the circuit "on" until the next button press. The other is switching the LED on and off rapidly, while a capacitor discharges to change the pulse ratio.

In the old days, you'd set it up on breadboard with a bunch of potentiometers while you dialed in the behavior, then start pulling pots, measuring them with the multitester and replacing them with fixed resistors. Which wouldn't quite work the same, because of all sorts of capacitive leakage between the various elements and other parasitic effects, so you'd pull those and randomly test a bracket of values until it worked again.



The last complex circuit I did in the pre-micro days, I didn't even use a 555. I used a hex inverter chip (which isn't an anti-magic shield, but six digital inverters in one package). You could make a pretty decent oscillator out of two or three inverters and the right discretes. If I remember correctly, it was a "power up" thing for a friend. Had a big rocket switch, and when you switched it a bar graph would crawl up to the top while a whine from a small speaker rose in pitch. All of that driven by a single beefy capacitor.



And these days, it is all micros. Because especially with the ATtinys, you are paying about as much per chip as you would be for the old integrated circuits -- or a pair of transistors, if you want to be really old-school -- and you tweak the timing in software instead of by trying out different resistors from your parts box. And they aren't any bigger; you can run the things basically naked, with their internal oscillator instead of adding an external crystal. And of course the program is stored on the same chip in non-volatile flash.

It is still dynamiting fish, but at least it is a bigger bucket. The example above, we had to fade the LED and latch the trigger button. That's just enough to make it easier to use a micro. The picture to the right, I soldered that up quick for a fire effect, using code to generate a random flicker on the red and green channels of the Cree (which, combined, make a more-or-less amber color).



And this episode reminded me once again, I really should get that PCB made. Ignore the RF for now; that's been distracting me too long. Just three MOSFETs, an ATtiny, some arbitrary ballast resistors, and a terminal or two to connect button or sensors. I just picked up some Chinese RGB's for under five bucks each, and I should be able to make the PCB for about another $5 even at the prices of a limited run at a fab house. So I could offer the complete kit for $15; a digitally controlled 3W RGB that could be user programmed and used in props or cosplay.

Wednesday, October 15, 2014

Lara Croft, Forklift Operator

Finally got around to finishing "Legend."* Right, a game from 2006. So sue me. The controls don't feel quite as fluid as "Underworld" (from two years later) but the "camera screw" (to use the TV Tropes term) isn't quite as bad, either.

The game is fun, even if the play is relatively short. On replay, with combat set to "easy," it finishes in under six hours. The puzzles are mostly environmental; as in, no real cases (outside of the Croft Mansion) where a lock is designed in the form of a complicated puzzle. More, this is figuring out how to get from spot to spot, or how to get weights to where they will sit on pressure plates, with various tricks using levers and blocks and grapples and so forth.

Still, several of these puzzles are up to "match the soup can" level; you need to do some odd combination of things which is not terribly intuitive. Such as, in one room, instead of searching the room you are supposed to swing a chandelier into a bell in order that the sound causes a crystal to break...

For me the highlight is really the over-the-top cheesy Arthurian Museum, with animatronic knights and a recorded announcer who pronounces all his "ye"'s. Which drives Alister (on headset) up the wall. And does a quite painless info-dump on Arthurian legend while it is at it.

During the museum, you get to drive a forklift. Which means at some point you are playing a video game in which you are driving around stacking crates with a forklift. It made me laugh. And was actually kind of fun. Also had one of the better bits of level design; down a flight of rickety stairs hidden behind a stack of crates is a stone mausoleum. Which contains as out-of-place artifacts two of the same crates. Which is to say; a very direct clue you are supposed to get the forklift down there, and use it to continue the adventure.

And it is most satisfying to drive the forklift right through the spinning blade traps, snapping them into little bits. And then you get to a portcullis gate, and Lara wonders aloud how she's going to raise the gate. And all I could think of is K9's response during "School Reunion," when Mickey (the other tin dog) asks if K9 has any high-tech tool to get them back into the locked school; "We are in a car."




And drat for continuity. My fanfic in progress arbitrarily admits "Legend" (but back-dates it to somewhere between 2001 and 2004) but I've been cagey about whether the first game (or, rather, the Core Dynamics re-make issued as Tomb Raider: Anniversary) is canonical.

Trouble is, now that I've finished Legend, it makes my fanfic a little out of character. Lara finishes that game with a new urgency to follow her father's clues and attempt to rescue her mother; in fact, the events of Underworld happen within a few months. She should not be kicking back on Malta and enquiring in a lazy way about a random Egyptian artifact.

Ah, well. It's just a fanfic.


* Here's the geeky bit.  Using wswine 1.7.28 engine, with "use wrappers quartzwm" checked and only "decorate windows" checked in the screen options. There's a flicker during mouse-over within the Steam wrapper, but once a game is started it runs without the dratted synch flicker and without crashing.

Sunday, October 12, 2014

Tool-Time

Finished my second day at the mill. As I've said, tooling can be hit-or-miss at TechShop, They do however have two boxes of end mills, and they were in good shape when I checked them out. So no problem at all milling the wider slots in my Suomi bolt.


The narrow slot that rides into the chamber was another matter. This is about 3/32 wide and a good .2" deep. Just getting the piece set up was interesting enough; I ended up with a stack of parallels to raise the bolt up sufficiently for quill clearance, with another one slipped in against the previously milled flat in order to achieve the 90-degree alignment I needed for the last slot.

I purchased a 1/16th end mill with a tapered shank (the only thing they had close to the right size) for this, and that was a bit scary. Put the RPMs up to 3800, used the paper trick to get the height. According to my very rough calculations, my cut depth should be around five thous per pass, and my feed rate "really slow."

There was no chatter or smoking coolant, so after a dozen passes I increased to ten thou per pass and a comfortable feed rate. Doing this by hand, and being careful to advance at a steady rate and not slam my tiny fragile end mill.

Widening the slot I was down to less than a quarter of the mill diameter so I could afford to stick a lot more length in there. That went quickly.

And the bolt fit. A little looser than I'd like, but I'm ready for welding now.




I also had a last grenade to run off -- slightly smaller dimensions overall so it would fit properly in an Airsoft shotgun. I thought I had a chunk of 7075 left but the small chips and lack of sound told me this was more of the 2011. (The main differences I noticed working 6061 is longer, better-formed chips, and a tendency to "sing" -- vibration noise.)

Unfortunately, the cross-feed gear was not engaging correctly -- it was as if it had a backlash of half an inch or so -- and I couldn't make accurate cuts (also, the runout was on the order of 60 thous). So I contacted staff, they took a look and said they'd probably have to order parts. And I waited for a user to clear the other lathe and hopped on that one when he was done.

And mucked up several of my cuts. I'd gotten too used to the mill or something and was consistently over-shooting. And I hope I cut it small enough so it will work.

The funny part is, though -- the guy on the other lathe? He was working from my Instructable!

Friday, October 10, 2014

TechShop Suggestions

I was reading a thread recently at Practical Machinist about how the forum members (most of them professional machinists with their own shops) feel about TechShop. And, as I'm getting ready for East Bay Mini Maker Faire I'm thinking again about talks I've heard there and through Park Day School about how kids learn technical skills.

TechShop is a successful business model and has an organizing philosophy I can get behind. That recognized, there are some stumbling blocks. Their funding covers the capital expense of the machines, but doesn't cover tooling, or sufficient skilled staff to achieve a high level of maintenance.

It is a similar problem many theaters (the building, not the organization) suffers; if you don't have a resident company, then maintenance and infrastructure suffers. No renter is going to put extra effort into cleaning up the mess from the last renter, and leaving the building better than it was when they got it. Even if they desired to do so, rental periods means they just aren't in there long enough.

TechShop, like a gym, has more members than could fit in the building at once. To control traffic at the larger, more expensive machines they use a reservation system. At some of the tools the maximum reserved block is a mere two hours. At lathe, mill, and CNC mill the period is four hours.

This is barely enough time to do the job and do a basic clean-up. It becomes difficult and frustrating when the tool was not restored properly or parts are broken or missing. I've spent as much as the first hour of a reserved block just tracking down the right wrenches,  finding tool holders in working condition, and assembling a dial caliper.

Which is not exactly a complaint; this is well within what should be expected for the fairly small membership fee, and it isn't in any way a game ender; I've been doing production work and even made a couple bucks profit off it under those same conditions.

In any case, one of the suggestions made at Practical Machining was to offer turn-key services.


Swap Board

I think, within the philosophy of TechShop and the Maker movement, this would work best in a more ad-hoc, bulletin-board sort of arrangement. There would be a central, organized marketplace for this, overseen and vetted by staff, with standards and guidelines. Approved projects could be taken by qualified members. This is a way that members who don't have the time or inclination to learn a new skill set just for one aspect of a larger project, and non-members who don't have the time or inclination to be a functioning member of TechShop, can get the kinds of small projects done that fall outside of the ability of professional shops to bid realistically on.

This might be developed into a sort of second tier of users; people who have specialized within certain skills, who get some degree of official support, and who as a result could be expected to take a more proprietary interest in maintaining certain pieces of equipment. In a perfect world, this pool would have a large intersection with staff and instructors; people who already are expected to spend a lot of time at TechShop and of course already maintaining the tools.

Among the many risks is this risks locking out the average users; less access to some of the tools with large blocks set aside for the turn-key bids, possibly more personal or "staff only" tools that are kept locked away from ordinary users. Which is all rather in opposition to the philosophy of TechShop.



There very much is a need for this. There are a lot of artists, in particular, who find themselves needing a weld or something and the number of hoops they need to jump through to do it themselves is daunting. And they lack two things to go to the existing infrastructure of small shops; the money to make it worth the small shop's overhead, and the ability to speak the language.

They need a sort of super-interpreter. They need someone who does what a good Technical Director does to a Scenic Designer; to translate ideas into materials. Someone who can plan a weldment, generate g-code, CAD up a shape, or even just understand drill sizes.

In the real world, this is a junction full of rough edges and, thus, friction (often resulting in heat!) In the prospect of a TechShop member accepting a task, they would be in many (if not most) cases accepting that they will be spending a lot of non-machine time planning, communicating, and teaching.

My thought is this is something some of us would enjoy. Given the right situation (stable finances, time to work), someone could get a lot of enjoyment in doing this kind of heavy lifting for an artist or small business person who thinks they want a weld done but doesn't understand enough about the materials they are working with to understand the potential problems.

Somewhere, somewhere in there is, I feel, a way to fold in the TechShop Instructors, the paid staff, the need for better shop maintenance, the pooling of skills that is part of Make and TechShop, and the marketplace of small-volume manufacturing that a small agile shop filled with CNC equipment can do.



Production Machines

Right. Enough on that.

I'd like to see two other things. The Tormach CNC is doing okay right now but the kinds of projects people bring to it require larger blocks than can be reserved. The result is at least one start-up that got memberships for half their staff and tag-teamed the machine, keeping it tied up for weeks on end.

Right now, the Trotech Laser is the only machine set up on a production basis. Anyone can come in and use the Epilog lasers for no additional charge, with only a two-hour maximum reservation. But if you have heavy production work, you reserve the more powerful Trotech and you pay an additional hourly fee.

I think we should do that for the Tormach. And possibly a few other tools as well. There are I believe enough users who would pay the additional hourly to allow longer machine sessions that you could capitalize the machine for it.


Outreach Class

In addition, I love TechShop and all, but I haven't been afraid of learning new tools since my first claw hammer. And not everyone is like that. I think a lot of people are scared off. I like that the TechShop philosophy is that anyone can learn, and each SBU is based on the assumption that six people off the street can successfully run a mill after just a few hours of familiarization. That's wonderful.

But there are a lot of people that look at all this heavy metal and complicated bits of machine and whirring computers, and hear all this chatter about feeds and speeds and g-code and TiAIN coatings and it is too much for them to handle.

I don't know what a truly introductory, get-your-feet-wet-in-the-wading-pool class would look like. I don't know how you detox it from the Burning Man, heavy industry, skilled-and-proud atmosphere. Maybe offer it somewhere outside the facility?

But there are kids and adults and artists and all sorts of people who don't believe they can learn this stuff, and we need to keep reaching outside of our own community and help them.


Milling around?

No, lathing a round. Milling a flat.

Used the (manual) milling machine today for the first time since my introductory class. Didn't quite finish all the cuts I need on the fake bolt for the Suomi -- so I probably won't be able to do the first trial weld tomorrow.

But I did use the mill successfully, without breaking anything. Now back to the books to refresh my memory on climb cutting, speeds and feeds, correct RPM for the edge-finder, etc.


The fake bolt is the top piece. The bottom is turned to the outside diameter of the barrel and pegs into the face of the bolt during the welding operation to keep the lugs aligned. Later I may replace it with a piece turned to the inside dimension of the barrel for more barrel support, but I might have to replace the turned peg with a steel rod for strength.



I also got a piece of 1/4" plate. I just might be able to do a quick run on the Tormach CNC mill -- the biggest problem right now, though, is that the desk is out of ball nose end mills. So I could do something that was a cut-out, but curved 3D shapes are not going to work well. Not until either they restock or I can order my own from McMaster-Carr.


Thursday, October 9, 2014

Practical Magic

More thoughts in re magic and technology.

Except for the fact that no wizards or gods were involved, something like a Damascus Steel blade functions a lot like a +2 sword. That is, it is measurably different -- better -- that most swords.

But it isn't a game changer. A +2 sword is just a little better. It doesn't have powers that go completely beyond sword-like. This is an artifact of game balance. Mythology can afford named weapons like Caladbolg or Kusanagi or the Odin-spear, lopping the tops of hills and killing armies in a single blow. Dungeons & Dragons makes a lot more use of the lesser, more common artifacts.

There were several ancient steels that owed their strength to what was essentially chance and secret incantations. Methods arrived at empirically and passed down verbally as closely-protected trade secrets, and luck of the draw in natural elemental ratios in the ore they were using. What gave the strength to a Masumune, or the tone to a Stradivarius, was unknowable at the time. It depends on subtleties in microstructure that there simply weren't instruments -- or the metallurgical understanding -- to read.



Anyhow, magic meets rationality, and at the moment rationality is winning. There's a strong trend in fantasy literature for systematic magic; magic that can be categorized and analyzed and basically behaves like another science. The two major breaks that occur to me at the moment are from the real world; one is the believers in psychic phenomena who have declared that (handily) scientific methods and the rational mindset itself is toxic to magic (thus neatly making it impossible to study).

The other is in several threads of religious belief. Basically, that god are above scientific laws, and simply redefine the universe as necessary to keep from being discovered by logic.

Oddly, although the first shows up in some works of fantasy -- usually as the reciprocal, with technology that would otherwise swiftly tilt the playing field conveniently ceasing to work within the Hogwarts the author desires to establish -- the more interesting explanation, that magic is a gift of the gods and the gods are, at the very least, free-willed and hence innately unpredictable, rarely utilized.

But I think there is another option a writer might take today. I think we have entered a cusp where basic logic and simplistic scientific method is incapable of pushing the boundaries of understanding much further. We are discovering more and more regions of knowledge, such as protein folding, that are so complex it is impossible for a human mind to properly comprehend and work with them.

In something as basic as the thermal flow and distribution inside a Dreamliner, calculation is impossible. Direct modeling is impossible (at least to the degree desired). Calculus even can not sum up all the variables properly. Only statistical methods and the weird brute-force fractal attack of finite-element-analysis (run on a supercomputer) can pull out the useful data needed by the engineers.

I am not saying we are reaching a point where things become intrinsically un-knowable. I am saying they become impossible to know in the more direct ways we've learned to know things. Back a handful of decades, computer vision was considered a trivial problem, artificial intelligence just around the corner, genetics was viewed as a simple code that once cracked would be an open book.

Now we are realizing our models were simplistic. Grotesquely so. There is no simple key to the genetic code, and machine vision is still stumbling. That said, even the second-hand ways of knowing; the rules of thumb, empirical findings, statistical measurements, FEA simulations, etc., can still be easily leveraged. You don't have to be able to understand how to program speech recognition in order to plug in a chip that already contains fairly efficient algorithms at doing so.

Which brings us back to magic spells and black boxes, as of my previous rambling essay on the subject.

So, yes, I think a writer might be able to defend a magic that could not be analyzed, reproduced, and mass-produced. Not because it was intrinsically impossible, but only because it had such interdependent complexities that modeling it properly was an incredibly difficult task. At some point, of course, the trade-offs work against magic. Either magic is so weak it doesn't matter that "technology" can't understand it. Or magic is so strong that technology is forced to spend the money and time necessary to tame it.

Which might make an interesting setting for a story, at that.

Wednesday, October 8, 2014

Props Progress

Suomi --

The first pieces are cleaned and ready for welding. I've found a wire welder at one of my theaters that I can use. Not neat or strong, but okay for filling; techies often describe them as "hot gluing with metal."

What's going to make it more plausible is the permanent jig. The piece of aluminium up there is standing in for a fake bolt I'm turning out of 1.25" aluminium. That will hold all the pieces together for welding and remain in place as part of the finished prop. As an additional advantage, it will look like a closed bolt when viewed through the ejection port.


Jubal Early --

Finding the correct dimensions for the frame was a matter of successive approximation. First I stuffed paper in there and used a graphite pencil to rub the edges. Then transferred that to foam-core. Adjusted the foam-core and transferred to MDF.

The last cut was in 3/8" basswood, which appears to be slightly wider than is correct. A lot of sanding and adjustment, and I finally had a mock frame that the Pachmayr grips fit properly.



Next is scanning or photographing. And it is unfortunately obvious that the grip other people are claiming is correct, is slightly different.

But I might blame Pachmayr. They are rather inconsistent about keeping the same names across different frames. (The different shapes visible here are not an artifact of that. That's an accurate depiction of the cut-out void inside the grip itself).

However, all of this does underline how uncertain my references are. I am sure I can get the final gun to look very close to the screen shots. But I am equally sure that the first prototype is going to involve at least some filing and sanding (and translating corrections back to the original CAD files.)

It's also not cheap. The top part is a slab of aluminium a full inch deep. A nice working size from Online Metals comes out to thirty bucks. So adding the stock for the rest of it, brass tubes and al, over a hundred bucks of metal and hardware. Then add the cost of how many end mills I may be destroying in order to cut that much.

On the other hand, an all-metal PPG is currently selling for over $900. So I think I can probably do this economically enough to be worthwhile.

Rambling Rex

Charlie Stross recently posted on his blog one of the ways in which modern technology resembles old-school fantasy magic. In one sense this is a trivial observation. Arthur C. Clarke, Florence Ambrose, and Agatha Hetrodyne have previously spun riffs on this.

But in another sense the insight is fresh. Modern technological artifacts are understandable, but very nearly un-comprehendible. And hand in hand with this is an increasing lack of user serviceability.

This is something the Maker movement has confronted. So much of our daily infrastructure is presented as inherently indecipherable. Consumer electronics -- both by design and as a natural outgrowth of the technologies of production -- are intended to be seen by the end-user as black boxes. The clearest point of this evolution, to me, is when Apple began putting their cases together with an extra-long torx key; a tool most end users would not have in their tool box.

Because of competition and industrial espionage, many pieces of equipment have the labels scraped off the chips, or the chips potted, to make reverse engineering harder. Now, there are people who have gone all the way down to analyzing micrographs to reverse-engineer a chip down to the transistor level. But for many of us, the need for increasingly specialized tools and specialized knowledge means even such hobbies as backyard mechanic are falling by the wayside.

So, practically, a great many things end up being treated as black boxes. Or, as Charlie put it, the kind of easy-to-use, easy-to-understand magic artifacts in a D&D game. Add to this a few layers of marketing-driven obfuscation, and the result is an end-user who expects their smart phone to work, transparently, all the time -- who is not expected to (when not actively prevented from) knowing what the real limitations of the technology and underlying science might be.

The place where it hits people like me, often enough, is when end-users extrapolate unfairly. They don't understand why the sound in an acoustically complex auditorium seating six hundred people can't sound just like the headphones on their iPod. And the reason is that so much is being done behind the curtain; automatic gain control, EQ curves tailored to in-ear transducers, leverage of psychoacoustic effects, etc., etc.

In short, the black box makes it look simpler than it is. And the nature of that black box makes it extremely difficult -- and, in practical terms, usually non-productive -- to try to open the lid.



Hacker mentality is about prying open the boxes whenever possible. It is about sufficient comprehension of the magic (and often leads to trickles of leaking Magic Smoke. Part of the Maker movement is a similar unpacking of the physical layer of modern technologies. There is something empowering about making your own injection-molded part or graphite-reinforced frame.

Or machining. I may be reading things that aren't there, but I think there is a healthy self-skepticism in the Maker movement. That Makers are aware that their knowledge is incomplete, even dangerously incomplete.

I am sitting here looking at a chunk of aluminium stock. It has markings on one end left over from the lathing I did to the rest of the piece. But I think if it in terms of one layer of archaeological investigation. One peel of the onion skin. How curves get into a piece of stock are not a mystery to me any more. More than intellectual knowledge, I have direct physical experience in cutting those curves on a lathe.

Into the next layer, I have some sense of the way different alloys are available, what an alloy is, what the properties are. I've direct lathing experience with a couple of different alloys, and that gives me a framework to at least have some sense of understanding what "corrosion resistance" and "machine-ability" mean in practical terms.

But even here -- when I order metal online, I sometimes get copies of the "MTR's and Certs of Conformance." Which I can only barely read, and have almost no understanding of how they are used. I am well aware that there are aspects to cutting a shape out of aluminium that are well and beyond my knowledge base.

And that's just cutting. Someone actually made that alloy. Someone mined that bauxite (or whatever). There are so many steps, so many associated processes, so many different sciences that were involved at some point in the bringing of that chunk of stock to where I can stick it in a lathe, it would be a lifetime of study to properly understand them all.

One word I use for this is footprint. Technologies have a footprint. The more complex a technology, the larger the required ecosystem.

If this is true of a pencil, imagine how much more true it is of an iPad.

Well, that is a bit unfair. We don't have to understand gravity in order to model it. We don't need to track aluminium all the way back to stellar nucleosynthesis in order to be able to lathe it properly. We don't have to be Thomas Thwaites, who decided to make a toaster from scratch. (Maybe he doesn't like Apple Pie?)

But it can be said that no one person can build an iPad. And that no one person designed it. It isn't quite the same as saying no one person could design one, but it is quite unlikely any one person would have the range of specialties necessary. And besides, that's just not how industry works. There are project leaders, there are even visionaries, but no one person is tasked with personally doing all the grunt work. Heck, even composers have been using copyists and arrangers for hundreds of years.

This means the artifacts of our creation are one step divorced from our direct design. They arise in collaboration, by committee, using already-complex existing parts in the growing libraries, and on the back of trends and existing standards and back-compatibility and inherited (yet unexamined) design assumptions.



Fortunately, you don't have to recreate an artifact of the modern world from scratch. You just have to understand the real black box -- the essential behavior, stripped of the masking environment -- in order to re-purpose it.

You don't need to have been the original creator of the spells "Summon Monster" and "Displace Spell" to realize you could combine them and decoy an attacker into a nearby wall. And you don't need to identify all the chips inside a toy in order to circuit-bend it into a musical instrument.

So if we accept what Charlie is saying, the implication is clear. Hackers are munchkins.

Poppins_01: Concept

This is probably not the best show for it.

But I have been meaning to document a sound design from the very top, through the process of development, and this is the next show I'm on contract for. This makes the first post in that series.




Some shows speak clearly; right away, from the first moment, you know what you want to do with them. Poppins, alas, is not one of those shows.

My first flush of a concept was to do something clever with the background sounds. I've gotten really tired (after doing Oliver!, Little Princess, and I can't think of how many other shows) of generic period London background sounds. And I've been thinking in terms of found sounds again, particularly since reading about the creation of the music for Tomb Raider 2013.

So it occurred to me to create these soundscapes from smaller scraps than usual. Rather than a horse and carriage from my usual libraries, make a more impressionistic horse with just a pair of hooves -- played into my keyboard and constructed from sounds that did not originate from an actual horse.

That way, I could change the character of the soundscape; contrast the dreary London Mr. Banks has been cowed to expect out of life, with with fantastical and colorful London of "Jolly Holiday." The same "clip clop" could change from a leaden, muffled rock-on-rock sound to a lighter almost bell-like hit in the same rhythms.




But I don't know if this will work after all.

Like a lot of shows, I've attended production meetings and listened to what the other designers are thinking of (sets and costumes usually bring images to show off as well). I've read the script at least once. I've been listening to a Broadway production recording, getting familiar with the music. But last week we finally had the first read-through with the full cast.

This is when the whole cast is introduced, and they and the design team all sit around a rehearsal room with scripts in hand and a piano, and we stumble our way through the entire production. So we hear all the dialog, and some attempt at the music.

And something struck me in this meeting that then, the next time I listened to the Broadway recording, was extremely clear. This is a show about silences.

There are many moments where there is just a thin melodic line suspended in air over the rooftops. And I need that open-ness. This is a place where as a sound designer I have multiple hats, but it all comes down eventually to me sitting at that board in front of an audience trying to pretend like I am at a dubbing stage for a motion picture; folding in dialog, stage noise, live orchestra, and sound effects into one complete picture.

What I am hearing in the score is that this complete picture requires empty space. Background sound effects are going to clash.

In other places, it moves FAST. The orchestration is busy, the singing is busy (lots of multi-part harmony and curse-of-Meredith Wilson stuff). There's not a lot of sonic space anywhere for passing carriages or the bells of St. Paul's or birds (fed or not.)



It may be that the open texture I desire is possible with the non-realistic sounds I was thinking of earlier. But I still don't have a grasp of how realistic and grounded the other departments are going. It can clash if the sound effects are non-realistic and the scenery is ultra-realistic! Or, sometimes, it can work. But that's a sense, a gestalt, that doesn't come until late in the production process.

I started "Wonka" with found sounds, and after listening to dress rehearsals with the full orchestra, did some hasty surgery to move as much as I could towards synth-based sounds instead. Those fit the total picture better -- but that picture from how the orchestra played when the actors were on mic, how the choreography looked on the painted set; that picture simply wasn't clear enough back when I was creating the original sounds.

In another week there is a full designer's run, for Poppins, followed by a sitzprobe. The former is when the cast stumbles through the entire show with full blocking (and without scripts in their hands). The latter is when the orchestra meets the cast for the first time. Usually, there will have been at least one (but often not more than one!) meeting between the orchestra members to rehearse in isolation. At the Sitz, they see if they can coordinate with the actual singers. A lot of Sitz ends up being the orchestra working out internally how they are going to make the pick-ups work. But it is also the first time I as designer get any real sense of how the orchestra is approaching the score.

I think orchestras don't think in this larger sense. For them -- to be more precise, for the Music Director -- there is a wad of music on the page that has to be gotten through, in a way that will support what the singers are doing and that doesn't sound too grotesque. Their effort is spent devising how to handle the harmonies and inner voices and little motifs that the composer called for, without (usually) anywhere near the assets the composer was hoping they'd have.

There isn't effort left over. But even if there was, most music directors don't think in the same terms as a mixer, or even as an arranger. They are all about "The score needs a horn, we can't afford a horn, the oboe can play that line so here's the oboe." What that choice is going to sound like in the physical space is not their affair. (And that's when they aren't actively sabotaging the mix by turning up their personal amplifiers so they can hear each other and get through the music as written...sacrificing en route the audience's ability to hear the music properly!)

In short, I have no way of telling whether I'm going to have these thin, delicate textures, or wether it is going to be a heavy, thick texture of over-amped piano and aggressive drummer and a synth trying and failing to pretend it is a string section.

Or something different and wonderful, like the punk rock band we had doing Pirates of Penzance.



Also in effects, there are effects that accompany various practical effects; falling plates, magical cakes, flying, whatever. I have tremendous trouble creating these before I have been able to see the effect itself. In Shrek, I am very happy with a few of the sounds, like the bird and the deer -- and those I saw the prop before I finished the sound. I was less happy with the dragon. The dragon sounds, to my ear, never matched the style and size and color and choreography of the final props -- props that didn't get delivered until opening weekend.

With three weeks to go, I'm not worried yet. But it is past time I made a rough cue list for those spot effects anyhow.

Monday, October 6, 2014

Late Early

I need to move the Jubal Early further back. Too many other projects with firmer due dates, and a bit too broke this month for 1/2" slabs of aluminium plate, 1/4" ballnose endmills, and a renewal on my TechShop membership.


Before I take a break on it, though, I want to get the model up to a show-off point.

The next step, though, is going slow. Like several recent props, the Jubal Early was built to use an available after-market grip. From research, specifically the Pachmayr Decelerator, in Smith&Wesson N-frame.

Trouble is, I can't find a proper scaled drawing of the actual gun frame (not one that allows me to properly import the curves into the CAD file). And I'm unwilling to cut apart a $30 pair of grips in order to scan the interior.

So I'm in the process of cutting a mock-up frame out of MDF, which I'll fit by trial and error. When that is finally right, I'll scan that in with a scale reference, and then build the CAD to that.


Another thing that is going to take time. There are several decorative screws. I don't know what size they are. Since a lot of catalogs don't bother mentioning the size of the head of a screw, I may have to purchase a selection from different manufacturers and measure them against the reference photographs until I have the right ones.

So even getting this up to the point of machining is going to take a while. The main body alone (seen above) needs to be flipped so I can cut from both sides, and then will probably need several manual machining processes; tapping screw holes, doing the through drill for barrel, and probably cutting a deep slot for the other parts to fit inside.

And as much as I am liking Fusion 360 so far, I think the CAM options on it may be too primitive for this. I didn't even see (on the one tutorial I've watched so far) anything on how to create support tabs. At the very least, I'm going to be pushing this through Cut3d (TechShop has licensed copies of that running on their machines), and I might even want to port to Inventor CAD for the final steps of prepping to machine.

I have to say; the Babylon5 PPG is looking easier and easier, relative to this one. The simple look of this thing masks that most of these cuts and rounded edges are tough to achieve in a controlled manner. This is an easy prop to hack out of a couple chunks of wood. It is a tough prop to machine to a high degree of polish out of proper metals.


Sunday, October 5, 2014

Gamergate

Insane that this is still going on.

The flag which the misogynists, haters, and general rowdies are still draping themselves with is "OMG, there's corruption in game journalism!"

In other breaking news, the sky is blue.

Have they LOOKED at the industry coverage? Ever? The majority of outlets "reviewing" AAA games are supported by the games they pretend to be able to make unbiased coverage of. Funny thing, but 98% of the games (totally unscientific impression) get favorable reviews. Sturgeon's Law alone says that is not possible. No industry can get that many right.

And the personal experience of too many individual gamers is broken mechanics, tired tropes, boring retreads, and buggy beta-ware. Any truly unbiased coverage would be taking a much harder look at that.

But, no, it is apparently all-important to raise a stink that an independent solo game designer had an affair with an editor (who has never reviewed one of her games in any case). Heck, I'd say, given the intense amount of environmental support -- the giant conventions, the sites crawling with advertising, the huge mainstream coverage -- that AAA games get, I'd be all for a little unfair advantage to the struggling independent game community. It needs it!

So, yeah. At the very, very best, the gaters are people who got stuck with the wrong end of an argument, and who are backing away trying to claim it was really all about "corruption in game journalism" when they were going full dudebro on a female game designer and denigrating her in every way that occurred to them.



I should really add -- games, like summer movies and paperback books, are commodities. You get bored with a game, even a good one, and you need to buy another. So it isn't per se irresponsible journalism to say "'Cauldron of Victory' is worth your $40," because it fills the need even if it is nothing more than "Medal of Honor" with orcs. 

I find this a lot more questionable industry practice when it comes to software, though. Every piece of software you adopt is a choice of potential lost productivity in retraining, migration of old files, compatibility with collaborators and the rest of your ecosystem.  And that is if if works flawlessly. It is in software -- Moore's Law be damned -- that a truly neutral review process would find no more than half of the titles actually worth trying.


Saturday, October 4, 2014

Con-fusion, pro-fusion

I am liking Autodesk's "Fusion 360" more and more. Finally figured out how to use proper spline curves (the tutorials were no help there). The major item in the "con" table right now is the cruddy documentation. I joined several other people on the comments page asking for a real manual. Watching someone zip through a tutorial without explaining how they made a selection, how they got to a tool or a sub-menu...(or that, in at least one place, what they were showing was actually impossible unless you'd turned off a preference in a totally different place, then re-started the application).

I'm up to about try 4 or 5 on the Jubal Early. Got through the basics of the upper receiver in Carrara but I'm dumping that. Doing it native in Fusion is going to make all the following stages much easier. And now that I understand their spline curves, I have half a chance of getting those weird curves right.




Yah; like a lot of things. The Fusion tutorials pretty much assume you've got a sketch, you want some curves, you'll adapt the design as you work. Replica prop making means someone shoved a chunk of stock into a band-saw, cleaned up the edges, handed to the actor -- and you are forced to try to replicate the result exactly, working from nothing but screen shots and dimensions off third-party casts pulled off a maybe/maybe-not screen-used original.

So most of the fancy modeling techniques in Fusion are of no use in roughing out that form. I need to reproduce those exact curves. And it has taken me most of a week to finally figure out where (and what) are the appropriate tools.




I'm going to laud the software in a moment, but first, one more con. 3d aps, for me, fall or shine based on how smooth it is to manipulate the view camera. Outside of Occulus Rift and a few other high-end bits of tech, we are viewing our 3d models and scenes through a 2d window. The only way to truly grasp the parts and their relationship is to be able to manipulate the view.

I turned off "use gestural control" again as it was useless. The only thing it did was dolly move and pivot. Pivot? Who puts pivot high on the view menu? There's no point at all in pivoting the view. Orbit is what you want. And orbit would have been so simple to code...! But, no, you get pivot. Which means you are constantly re-setting the resulting Dutch Angles. Oh, and the application breaks the GUI standard -- no space bar drag. Why? The spacebar isn't coded to anything else. They could have allowed spacebar drag. Instead you have to mouse over to a little widget, drag the viewport, mouse back up to the selection tools, go back to clicking. It breaks the flow.

This is why, sad to say, Carrara is still useful. Because you can navigate and select and manipulate practically simultaneously, just by holding down different combinations of command and control while you move the mouse.

In everything else, of course, Carrara basically sucks.

Fusion almost gets a pass for the slower viewport manipulation due to render tricks that make it vastly easier to see what the hell you are doing. It uses a translucent, fully-shaded object, with reference images (when you want them) floating on planes and also translucent.

But here's what else you get:

Fully parametric modeler. You specify lines and curves mathematically, meaning they are infinitely smooth, scaleable, and editable. It also has parametric methods to model along mirror planes, work on multiple clones simultaneously, generate topologically complex objects such as screw threads, and so forth.

It is an integrated environment. You can check dimensions on (mostly!) arbitrary points, generate fixed dimensions and even generate complete printable plans.

You can also check clearances, set standard clearances or interference fits, create and animate mechanical motion, and do a lot of other useful little functions for 3d print and CNC such as making a volume check.

It will also generate and edit CNC tool paths. I don't believe it slices for 3d printing, but oh well -- that's easy enough. Although I haven't gotten far enough to see if it can export in formats that are readable by standard slicers and other manufacturing prep software.

I also haven't gotten far enough to see how clean the meshes are if you export in OBJ format. So I may need to turn to Blender, still, for Poser product creation. But then, Fusion has as far as I can tell nothing resembling UV mapping tools.




So the gunsmithing continues. Prop-gun-smithing. I've cleaned enough of the slag from the Suomi receiver to start messing around with welding. Current plan is to install a fake bolt machined out of aluminium; that will hold the pieces aligned, and give it some extra strength. It will also fill the empty space visible through the ejection port.

(I'd be about willing to skip the welding at that point and go for Liquid Weld or similar. Except that I purchased a couple cans of Gun Blue and that's only going to look right on real metal).

A pair of Pachmayr grips arrived in the mail and I'm measuring them to fit the Jubal Early to them. That's going to be a little while yet; I have to finish the model, prep that for CNC, and order a rather pricey amount of metal (the Jubal's upper body is a slab-like 26mm thick.)

I've been posting images of the evolving 3d design for the "flash hider" of a "King of the Rocket Men" gun, which may possibly be turned in aluminium. If I get design approval I may be borrowing a (real!) luger P-08 in order to take detailed measurements.

And, yes, I still have the pirate pistol. When my head can't take learning new 3D software anymore, I'll switch gears and go back to learning how to parse sensor data in Processing.




Oh, yeah. Switching out my old desk chair for a piano bench is doing me a world of good. I'm able to spend a lot longer at the computer without pain in my legs. Getting rid of the chair, setting up a proper tool area in the kitchen and putting my keyboard controller out also makes the place feel a lot more spacious and clean and also is bringing my spirits up. Now I just have to reserve enough time to repair more audio gear and start creating sound effects for Poppins!

Thursday, October 2, 2014

I bought a prop today oh boy

Back when I was young I somehow managed to acquire the AMT "Star Trek Exploration Kit"; a plastic model kit for the phaser, tricorder, and communicator in roughly 2/3 scale. I actually put tiny buttons and LEDs in the communicator but it was a bit beyond me to stick a tone generator in the remaining space.

Later, at one of the big conventions, I picked up a 1/2 scale "Life Clock" in cast resin.

And that I think is it for actual television or movie replica props. Sure, I've collected a few patches and lanters and so forth here and there. But these were more generically historic replicas, not specific to any one movie or television show.

Which makes it odd that for the last couple of years I've been involved in the replica prop community, and have personally contributed twenty-six of the pulse rifle grenades from "Aliens" back to the community. And I'm now simultaneously in the planning stages of a Commando Cody ray gun, and the gun used by Jubal Early on the television show "Firefly."

Well, today, I joined the community properly. Entertainment Earth had TOS phasers on sale for $14 and I picked one up. (Considering the Pachy's alone for the Jubal Early cost $28 -- also on sale -- I don't think that's too much of a splurge!)

I also have a pot-metal flintlock pistol at the moment, to help me to generate sensor data for the "Trigger Finger" project. Another thing I really want to complete and get out there as a purchasable kit.



(Meanwhile I'm learning Fusion 360, I finally set up my Behringer 61-key controller and am trying to recover what little piano-playing skill I once had, and, oh yes, the first musical of this season is going into tech in another three weeks.)

Wednesday, October 1, 2014

Cold Fusion

Well...took three hours to get it to run at all. Apparently you need to start an account, then start Fusion, then go back to your account at which point it starts a cloud folder, then start Fusion again or....something...and eventually it will actually start.

And it is sluggish. The camera keeps moving randomly because the software is just too slow to keep up with itself and know if you are just moving your selection pointer to a new selection, or trying to move the camera. That's on a dual i7 with HD Graphics 4000. Not the greatest system, sure, but a mid-range laptop that's not much more than a year old. And if your general-distribution software doesn't run on that, you may want to rethink what you mean by "general distribution."

I'll keep messing with it, but I have models that need to be done NOW.


More findings. It is a little laggy, but the camera issue is because Fusion's code base apparently over-rides the Mac for reading trackpad input -- regardless of how the preferences in Fusion are set up. The Mac OS is of course tuned to the Powerbook...and Fusion is not. So if my thumb is within a half inch of the pad (aka is within my usual mouse position) Fusion will read it as a two-finger swipe instead of a cursor move. If, on the other hand, I use two fingers, Fusion will read that as a three-finger swipe for its purposes. 

What did I say a few posts ago about 3d softwares violating the GUI? The only thing I can say in Fusion's defense is after you've learned a completely new muscle memory, it appears you would be able to navigate the 3d workspace very efficiently. 

I still wish I could lock rotation. Rotation is essentially useless to me when I'm in an ortho view trying to snap to a reference image. There is a reason why rotation is usually harder to get to in most 3d navigation schemes; it is used so much more rarely. Alas, not in fusion. Since it reads any two fingers as a camera control, any shifting of those two fingers relative to each other is read as a command to rotate the view in plane.

I think, all in all, I prefer the Carrara and DAZ scheme where you implicitly hold down a key (usually control) to change the behavior from free-moving mouse to camera control.