Microchip doesn't appear to be storing OSCILLATOR ERROR values for 3224's?

[SpenceKonde]

Kind of a bummer, I gotta say...

At least the 1624;s had them,.

Anyone else playing with 2-series parts care to see what their sigrows look like? you can print the whole thing like so

volatile uint8_t * sigptr = (volatile uint8_t *) &SIGROW;
  sigptr = Serial.printHex(sigptr,16, ':');
  sigptr = Serial.printHex(sigptr,16, ':');
  sigptr = Serial.printHex(sigptr,16, ':');
  Serial.printHex(sigptr,16, ':');

output looks like
1E:95:28:30:43:4D:4E:32:43:AE:6C:4B:12:FF 3F:0E
06 FF:FF:FF:14 05 FF:FF:A2:05:1E:06:FF:FF:E0:F8
88:EC FF:FF:FF:FF FF:FF:AE:F4:E3:FC:1F:FF:FF:FF
FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:F1:BE

Bold text is where the osc error used to be. 2-series oscillators are so good that the values in there were like 0xFD (-3) to 0x03 (+3), and usually -1 to +1

Italics are values that vary between individual specimens and are not documented. They vary over a very narrow range: the first one is always pretty close to 0x40 (nominally 64?) the next one is 0C, 0D or 0E on all parts I've seen, and the other two are always 05, 06 or 07. No strong correlations between any of them,

Looks like they've decided their oscilators are good enough that they don;'t need to...?

[hmeijdam]

here you go, this comes from my 1626

1E:94:29:30:42:42:38:51:38:41:F8:27:44:FF:40:0E
06:FF:FF:FF:12:07:FF:FF:B6:04:34:04:FF:FF:E0:F8
8C:18:FF:FF:FF:FF:FF:FF:AE:F4:E3:FC:1F:FF:FF:FF
FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:8C:9B

and as a comparison this is from a 806

1E:93:24:32:50:32:51:4E:36:06:65:21:53:FF:42:0E
06:FF:FF:FF:10:07:1B:05:5B:05:9A:05:D9:05:EF:F8
95:02:04:07:FE:00:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF
FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF

and a 416

1E:92:21:51:50:51:50:4C:20:39:94:21:4A:FF:40:0E
06:FF:FF:FF:12:08:19:07:58:07:98:07:D7:07:EF:F8
93:01:FC:03:04:0A:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF
FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF:FF