Skip to content

Commit 03e519c

Browse files
committed
update chip status
1 parent 0102cb2 commit 03e519c

File tree

1 file changed

+3
-4
lines changed

1 file changed

+3
-4
lines changed

README.md

Lines changed: 3 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -29,12 +29,11 @@ Chip 2: Prototyping Chip, status: some cells over-SET, working otherwise, nothin
2929
Chip 3: FORMed all, programmed with first set of inputs (HRS: 70-90kOhm)... was somewhat working, but then we tried to push to 500kOhm and shorted a bunch of cells. Lesson learned: don't push past 100kOhm.
3030
Chip 4: FORMed every 8 cells, programmed with a simple debug vector, working but cell HRS is 70-90kOhm
3131
Chip 5: FORMed every 8 cells, programmed with a simple debug vector, HRS is >100kOhm, working
32-
Chip 6: FORMed all, programmed with first set of inputs (HRS: >100kOhm), working!!! But now shorted
33-
Chip 7: FORMed all, programmed with second batch of inputs (HRS: >100kOhm), TODO: reexamine.
34-
- Programming batch_0 vectors and tuning HRS
32+
Chip 6: FORMed all, programmed with first set of inputs (HRS: >100kOhm), working!!! Re-measured accuracy matrix vs. bias conditions.
33+
Chip 7: FORMed all, programmed with second batch of inputs (HRS: >100kOhm). Re-programmed to batch_0 vectors (lots of stuck-at error bits).
3534
Chip 8: FORMed all, programmed with a simple debug vector (same as 4), we think the sense amp might be broken
3635
Chip 9: prototyping
37-
Chip 10: FORMed all, programmed with batch_0 feature vectors (HRS >100kOhm)
36+
Chip 10: FORMed all, programmed with batch_0 feature vectors (HRS >100kOhm). Measured accuracy matrix vs. bias conditoins (consistent with Chip 6).
3837

3938

4039

0 commit comments

Comments
 (0)