The following reply in the email discussion between Peter van der Made and Trollis (alias Travis) should provide comfort to some even if one anonymous poster still has doubts:
From: PeterSent: Friday, May 17, 2019 11:55 AMTo: TravisSubject: RE: Your questionsTravis,
I cannot go into that level of detail. Eliminate the idea of a transistor for each synapse ( which does not work, especiallysince synapses are either 1bit, ternary, 2bit or 4bit) and think way out of the box. Socionext have our design and theyhave not told us its impossible. No only that, we have an FPGA and demos to prove our point. We are not disclosing oursecret sauce to anyone.
Best regards,
Peter
NB: This email was sent to Travis in full understanding by Peter van der Made that it would be released generally and most likely on this forum. If the statements made therein are untrue or misleading the personal financial consequences for the author would be significant if not catastrophic.
What Peter van der Made had to say
-
- There are more pages in this discussion • 23 more messages in this thread...
You’re viewing a single post only. To view the entire thread just sign in or Join Now (FREE)
Add BRN (ASX) to my watchlist
 (20min delay)
|
|||||
|
Last
16.5¢ |
Change
0.000(0.00%) |
Mkt cap ! $323.8M | |||
| Open | High | Low | Value | Volume |
| 16.5¢ | 17.0¢ | 16.0¢ | $842.4K | 5.117M |
Buyers (Bids)
| No. | Vol. | Price($) |
|---|---|---|
| 11 | 316637 | 16.5¢ |
Sellers (Offers)
| Price($) | Vol. | No. |
|---|---|---|
| 17.0¢ | 1386823 | 34 |
View Market Depth
| Last trade - 16.10pm 20/09/2024 (20 minute delay) ? |
| BRN (ASX) Chart |