Breaking News
View Comments: Newest First | Oldest First | Threaded View
Bert22306
User Rank
Author
Re: Proprietary protocols
Bert22306   5/16/2014 4:05:08 PM
NO RATINGS
That too, actually, or "almost." In our case, in 1999 or so, we had to develop an actively redundant link layer protocol, for high speed, high survivability networks. Lo and behold, in 2010, IEC 62439-3 is created: Parallel Redundancy Protocol. It's not identical to ours, and I much prefer our approach for a couple of reasons (nothing like developing your own to understand the advantages and disadvantages of other approaches you had to explore in the process), however overall, the way it works is the same. The differences are in flexibility of the link layer topology and in maximum speed the protocol can support.

antedeluvian
User Rank
Author
Re: Proprietary protocols
antedeluvian   5/16/2014 8:47:52 AM
NO RATINGS
Bert

Thanks for your input.

The fun part of the job is to keep track of the ongoing standards efforts, and then adopt (and adapt) the successful ones when it makes sense to do so, while still retaining compatibility

the thought occurs to me that on rare occasions this goes the other way, when a proprietary bus is standardised. I'll bet every standard of early serial bus started that way. GIven the structure of Modbus, you can just feel the evolution from a single purpose idea into the general purpose standard.

 

Anybody out there had their bus/protocol become a standard?

Bert22306
User Rank
Author
Proprietary protocols
Bert22306   5/14/2014 4:36:32 PM
NO RATINGS
Absolutely! This topic of rolling your own, vs. using an existing standard, came up in a recent article by Junko Yoshida as well.

We very much created our own protocols, and entire digital networks, before standard ones were good enough to do what we were building. The fun part of the job is to keep track of the ongoing standards efforts, and then adopt (and adapt) the successful ones when it makes sense to do so, while still retaining compatibility with the proprietary protocols that came before.

Ultimately, for our work, adopting standards allows you to improve the products more rapidly. Because you're leveraging off the work of thousands of smart designers all over the world. However that wasn't the situation at the beginning. For my work, this has been true for protocols and also for hardware.

Most Recent Comments
michigan0
 
SteveHarris0
 
realjjj
 
SteveHarris0
 
SteveHarris0
 
VicVat
 
Les_Slater
 
SSDWEM
 
witeken
Most Recent Messages
9/25/2016
4:48:30 PM
michigan0 Sang Kim First, 28nm bulk is in volume manufacturing for several years by the major semiconductor companies but not 28nm FDSOI today yet. Why not? Simply because unlike 28nm bulk the LDD(Lightly Doped Drain) to minimize hot carrier generation can't be implemented in 28nm FDSOI. Furthermore, hot carrier reliability becomes worse with scaling, That is the major reason why 28nm FDSOI is not manufacturable today and will not be. Second, how can you suppress the leakage currents from such ultra short 7nm due to the short channel effects? How thin SOI thickness is required to prevent punch-through of un-dopped 7nm FDSOI? Possibly less than 4nm. Depositing such an ultra thin film less then 4nm filum uniformly and reliably over 12" wafers at the manufacturing line is extremely difficult or not even manufacturable. If not manufacturable, the 7nm FDSOI debate is over!Third, what happens when hot carriers are generated near the drain at normal operation of 7nm FDSOI? Electrons go to the positively biased drain with no harm but where the holes to go? The holes can't go to the substrate because of the thin BOX layer. Some holes may become trapped at the BOX layer causing Vt shift. However, the vast majority of holes drift through the the un-dopped SOI channel toward the N+Source,...

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
Like Us on Facebook
EE Times on Twitter
EE Times Twitter Feed