| Thread Links | Date Links | ||||
|---|---|---|---|---|---|
| Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
I vote NO.I believe Motion 19.02 shows mid-rad can be introduced into IEEE 1788 without compromising the strengths or importance of explicit inf-sup formats.
In regards to Kaucher arithmetic, even Arnold admits: On 2010-09-16 Arnold Neumaier wrote:
Note that I am neither against midrad nor against Kaucher intervals. In particular, I have made constructive use of both in my work.
We also have been using Kaucher arithmetic constructively for over 5 years. Particularly for efficient computation of narrow bounds on Bezier, B-Spline and NURBS curves and surfaces, a Kaucher arithmetic processor will always be faster and more efficient than an interval processor that does not support the Kaucher arithmetic.
The reasons for this have already been examined, discussed and explained at length both in this forum and in the various white papers that have been on the IEEE 1788 website now for several years.
I suspect that if P1788 does not standardize Kaucher arithmetic, end-users will notice that non-conforming hardware which supports the Kaucher arithemtic will run programs faster than hardware from vendors with an "IEEE 1788" sticker on it and this will lead to questions why did P1788 not standardize the Kaucher arithmetic.
I believe the time for standardization of Kaucher arithmetic is ripe, if not alreay overdue.
I would also point out that Motion 12 has already been accepted, and is isomorphic to Kaucher addition.
Sincerely, Nate Hayes Sunfish Studio, LLCP.S. I also believe if this motion passes that P1788 needs to change the name of the standard, since clearly mid-rad and Kaucher arithmetic are "interval arithmetic," and it will be misleading for P1788 to persist in calling IEEE 1788 a standard for "interval arithmetic" if these aspects of interval arithmetic are not incorporated into the standard.
----- Original Message ----- From: "Ralph Baker Kearfott" <rbk@xxxxxxxxxxxx>
To: <owner-stds-1788@xxxxxxxxxxxxxxxxx> Cc: <stds-1788@xxxxxxxxxxxxxxxxx> Sent: Sunday, October 10, 2010 11:03 AM Subject: Motion P1788/0023.01:NoMidRad -- VOTING PERIOD BEGINS
P-1788, The motion discussion period has run out on October 5. Althoughthere are several other issues also presently past their discussion period,those are not closely related to this issue, and it is important that this issue be decided soon. Thus, I have decided to bring this issue to a vote separately. Here, let us agree that "support" in this motion means that operations on the object, possibly including accuracy and reproducibility requirements, are explicitly defined in the standard. Also, by "nonstandard intervals," let us agree that this means Kaucher arithmetic. (Otherwise, one might think "nonstandard" meant "anything not in the standard," something that would not make sense in this context.) The voting will continue until after Sunday, October 31, 2010, or until such time as the Voting Tabulator deems is appropriate. Juergen: Please post this information on the web page. Guillaume: Please record this in the minutes. The motion is as follows: ---------------------------------------------------- The standard shall not support a midrad interval format or nonstandard intervals, beyond providing conversion support, approximately to the extent specified in the Vienna Proposal. ---------------------------------------------------- Best regards, Baker -- --------------------------------------------------------------- R. Baker Kearfott, rbk@xxxxxxxxxxxxx (337) 482-5346 (fax) (337) 482-5270 (work) (337) 993-1827 (home) URL: http://interval.louisiana.edu/kearfott.html Department of Mathematics, University of Louisiana at Lafayette (Room 217 Maxim D. Doucet Hall, 1403 Johnston Street) Box 4-1010, Lafayette, LA 70504-1010, USA ---------------------------------------------------------------