Subject: Re: [ublas] Performance woes affecting ublas
From: Andrea Cassioli (cassioliandre_at_[hidden])
Date: 2010-05-12 03:04:15
I do agree with you!
If I'm not wrong, a couple of weeks ago someone post that a
documentation revision would be under go soon. I hope it will improve
the current situation.
On 5/12/10, Rui Maciel <rui.maciel_at_[hidden]> wrote:
> Andrea Cassioli wrote:
>> Hi Rui,
>> I must agree with you about the lack of performance of uBlas also for
>> small sparse matrices. I have experienced similar problems in
>> performing simple matrix-vector products, but in my case I could
>> rewrite my code to explictly perform the computation.
>> The timing you claim are indeed impressive, but I more fair comparison
>> should be done using uBlas in the proper way (something small
>> differences in coding result in great saving of time), but I do agree
>> that the documentation is an actual problem for the uBlas community.
> Yes, I agree that it would be better if it was possible to perform some
> benchmarks that could fairly and objectively compare this stuff. Yet, as I
> it, the main problem is that, even if ublas didn't crawled when compared to
> other packages, the documentation as it stands makes it a bit hard, if not
> practically impossible, to be able to write code that runs as it should.
> Rui Maciel
> ublas mailing list
> Sent to: cassioliandre_at_[hidden]
-- Andrea Cassioli