ridiculous advice "don't sort 9 (nor 99000, perhaps)
numbers; instead pad with throw-away values to 10^5 numbers. I hope my
result is not CPU-model specific, and you can observe something similar.
No issue in 8.07.
Best regards,
Vadim
On Fri, Jan 27, 2023 at 6:18 PM Henry Rich wrote:
f workers are busy and none of them slacks". Thank
you very much for the detailed explanation, this cutting into very small
pieces is new to me, I'll now see if I can implement it in practice.
Sorting is only part of a larger whole; it's just that I noticed
interesting results whil
ork. But perhaps it could with certain data, CPU model,
number of cores? So my question is if you could confirm that it (slower
than I expected speed with numerical sort in threads) is neither J issue,
nor 't.', nor '/:~'. Sorry if I wasted your time.
Best regards,
nds of different constructs investigating why multi-threaded
operation (with "t.") on subarrays is so much slower than expected,
although it's perhaps totally unrelated to what's discovered above.
Best regards,
Vadim
--
For information about J forums see http://www.jsoftware.com/forums.htm
cted a minimum stack size of 16mb. Try Building your front end with
> that stack size.
>
> On Sun, Mar 6, 2022, 6:35 AM vadim wrote:
>
> > I'm considering to move a working program (which loads j.dll) from J
> > version 8.07 to current 9.03, but JDo fails to execute any sen
JDoType JDo;
void *j;
int result;
//hinstLib = LoadLibrary("C:\\Users\\vadim\\j64-807\\bin\\j.dll");
hinstLib = LoadLibrary("C:\\Users\\vadim\\j903\\bin\\j.dll");
JInit = (JInitType) GetProcAddress(hinstLib, "JInit");
JDo = (JDoType) GetProcAd
other hand, I appreciate that your solutions are very general,
thank you for shedding the light on correct path. E.g. "P" can be
easily replaced to test for unordered subsets instead. Then another
dilemma, whether repeated part numbers are allowed on one line...
Best regards,
Vadim
On Sat, Jul
> a of A is maximal in X (there is no item of X larger than it) _if and only
> if_ it is maximal in A and none of the maximal elements of B are larger than
> it.
>
> fastmax is so quick on syms1000 simply because of the amount of repetition.
> My guess is that recmax would comp
omparing most elements to all the others.
>
> Might come back with some more ideas.
>
> Cheers!
> Louis
>
> > On 28 Jun 2019, at 20:20, Raul Miller wrote:
> >
> > Does order of the substrings matter? (In other words, would it be ok to
> > treat 'B002|A0
Example: given a file (a string), where each line is list of part
numbers with separators, how to exclude lines which are substrings of
other lines (cut on separators)? E.g. (first line is to be excluded):
z =: 0 : 0
A001|B002
C003|A001|B002
B002|A001
C003|D004|A001
E005|F006
D004|C003
)
[li
10 matches
Mail list logo
