On 1/29/12 5:19 PM, Paul Bowyer wrote:
> On 01/24/2012 05:17 PM, Robert Dodier wrote:
>> On 1/24/12, Paul Bowyer<pbowyer at olynet.com> wrote:
>>
>>> So far my lisp programming is to the point where I can write simple
>>> programs around CLX, but I haven't tried anything as advanced as maxima.
>>> I think it would be well beyond my abilities.
>>>
>>> If I discover anything in my stumblings with maxima, I'll post them, but
>>> don't hold your breath waiting...
>> OK, well, Maxima is a big, messy program but in some ways it is not
>> too terribly sophisticated. I think it's likely that you can readily
>> learn enough about Lisp& Maxima to make useful contributions.
>>
>> best,
>>
>> Robert Dodier
>>
> Robert:
>
> I tried updating CCL and rebuilding maxima, but it made no difference
> with the testsuite errors. The CCL maxima runtime still fails rtest15,
> problems 37, 193, and 196.
>
> build_info ();
> Maxima version: 5.26.0_26_gc4216e7
> Maxima build date: 18:32 1/28/2012
> Host type: i686-pc-linux-gnu
> Lisp implementation type: Clozure Common Lisp
> Lisp implementation version: Version 1.7-r15188M (LinuxX8632)
I think Barton mentioned a little while ago that the errors only occur
with the 32-bit version of CCL. Everything works fine with the 64-bit
version.
>
>
> Result:
> Polynomial quotient is not exact
Ugh. Of all the bugs to start with, this is probably one of the most
difficult but also one with far reaching consequences. No one has found
a case that was simple enough to debug and no one has figured out
exactly how it all works. (I think.)
If you're new, maybe this is not the place to begin. On the other hand,
maybe a newbie can learn enough and has the right point of view to
figure it all out. :-) That would be fantastic!
Ray