#183

omg. this is absolutely amazing!

#184

I definitely plan on open sourcing my design, however, I’m not 100% sure how I want to go about that. For example, I thought about trying to sell a batch of them (say 5 or so), and after they sold out release the files. I dunno.

Also: these are very expensive to print through a commercial service. I just priced out the printing cost on Shapeways and it is roughly $275 USD to print the top and bottom parts of the case using their cheapest materials. If I were to print a batch I reckon I could do it for cheaper (and probably better too given that I designed it).

Also: my current printer is not big enough to print these as single pieces, so I ended up printing several parts and glued them together. If there were enough people interested in buying a case, I might take the jump and invest in a larger printer (using the proceeds to finance the whole operation).

Here’s another picture with some of the many prototypes in the background:

IMG_20200203_120101331
IMG_20200203_1201013313072×4096 2.58 MB

13 Likes
#185

Looks nice and I was considering doing something similar but was put off by having to do it in sections as my printer has max dimensions of 250 x 200mm.

I am considering other options and I wondered if people have ideas on fastening the top layer to the body without having visible bolts. Because aesthetics. I have an old 40h kit which I made with an aluminium top layer, and it just stays in place with gravity and a snug fit in the surrounding wood case, but I would prefer a more robust design. Do the ‘official’ grids have screw holes in the underside of the top surface and bolts from below?

1 Like
#186

Mine has screw holes on the underside that connect to the top plate. The top plate is thicker than it looks (it drops down after the lip of the case), allowing enough depth for adding screw holes.

1 Like
#187

…That inevitable point when a DIY project becomes as expensive (or more expensive) as “the real thing” :laughing:

2 Likes
#188

Still a fair bit cheaper, even so: $150 for the trellis parts, and say $150-200 for the case (if I were to do a batch): so $350 total. Last I checked grids were $700.

3 Likes
#189

Talking more about the trial and error R&D factor - and exaggerating for emphasis. :wink:

And then… time. The acrylic case I did takes like an hour just to peel the paper backing off all the parts.

6 Likes
#190

Yes, but you have a 256 as a result! Sure, use cases were always limited on a 256 (even when they were produced), but got dang is it cool! I would almost want to build one just to load up old scripts and go nuts!

1 Like
#191

Am I correct in taking this to mean “other scripts will work as though it were a 128” (i.e., only the top half), or are there scripts which outright fail when talking to a 256?

For me, just being able to run loom and flin on the Really Big Grid might be worth it, but if there are actual errors from other scripts…

#192

Yeah. Everything else should just use the top half and work normally (like cheat codes seemed ok)

Scripts will likely have things hard coded to 16x8 (As that’s “the norm”). But they can probably be modified to 16x16… but maybe that’s less efficient (?) if you don’t have a bigger grid.

I’ll do some experiments when I have time.

3 Likes
#193

Ugh. I hadn’t even considered this… not looking forward to it :<

EDIT:
Forgot to say

Maybe it’s easier to read on a varibright monome grid, but the lights which happen when loom’s “threads” collide are really evident on the Launchpad Pro using “rainbow” mode (like, a different color entirely), whereas they disappear into the glare on the neotrellinome in monochrome mode… but perhaps this speaks to a need for (especially) gamma fine-tuning.

1 Like
#194

The way I did it (smooth color variation AND intensity) should work with looms. I’ll try it out tonight.

#195

Nice! Are you pushing a PR to @okyeron’s repo, or do you have a fork?

#196

Will you add the hex in github for the 256?

#197

not right now.

the only changes are

#define NUM_ROWS 16 // DIM_Y number of rows of keys down
#define NUM_COLS 16 // DIM_X number of columns of keys across

and

// NeoTrellis setup
Adafruit_NeoTrellis trellis_array[NUM_ROWS / 4][NUM_COLS / 4] = {
  { Adafruit_NeoTrellis(0x33), Adafruit_NeoTrellis(0x31), Adafruit_NeoTrellis(0x2F), Adafruit_NeoTrellis(0x2E)}, // top row
  { Adafruit_NeoTrellis(0x35), Adafruit_NeoTrellis(0x39), Adafruit_NeoTrellis(0x3F), Adafruit_NeoTrellis(0x37)}, // bottom row
  { Adafruit_NeoTrellis(0x43), Adafruit_NeoTrellis(0x41), Adafruit_NeoTrellis(0x36), Adafruit_NeoTrellis(0x3E)}, 
  { Adafruit_NeoTrellis(0x45), Adafruit_NeoTrellis(0x49), Adafruit_NeoTrellis(0x4d), Adafruit_NeoTrellis(0x47)} 
};

my jumpers/hex-values were kinda chosen at random - I’d probably want to clean that up to something more sensible.

1 Like
#198

Tested - Loom works on full 16x16.

If a script is written appropriately with something like this (from loom) it should just work.

    grid_w = grid_device.cols
    grid_h = grid_device.rows

FWIW - I don’t think Flin has been updated to norns 2 syntax
flin here: Norns 2.0.1

1 Like
#199

For now, I put it on my github - consider this a beta test - I got it working a couple days ago. If people like it, and Steven thinks it adds something it would be better to have on his page. I might add a “press all four corner buttons to re-select a palette” feature, which would be handy.

And yes, it works wonderfully with Looms:

12 Likes
#200

NIIIICE!

Can you give more details on how you selected the colors? I’m thinking about trying out a “grayscale” palette…

#201

I just used a color selector app in windows, took a screenshot of the palettes, and tried to select 15 colors (plus 0,0,0) with the eyedropper, and typed in the RGB values. Very crude, could use some refinement. If you want true greyscale, replace one of my tables with 255,255,255 for every value. The gamma intensity vector then gives greyscale. Or just use Steven’s original code.

3 Likes
#202

@okyeron
Appologies, I completely forgot I was going to post this.

heres my max patch for testing USB decoding/LED driver throughput when developing vb_driver.

dbd.maxpat (18.3 KB)

2 Likes