On 16-May-07, at 3:04 PM, [EMAIL PROTECTED] wrote:

> On May 16, 2007, at 17:44 UTC, Frank Condello wrote:
>
>> Daniel's profiling a loop that uses direct Ptr access - there are no
>> function calls.
>
> What?  Is Ptr.SingleValue(index) not a function call?

Ptr.SingleValue is a function call, but the OP's code uses  
Ptr.Single, which isn't a function call.

>> The report evaluation suggests he wasn't testing in release builds
>> which is why he was seeing such a big discrepancy.
>
> Ah, we should have asked that right off -- there is indeed a fair
> amount of overhead in debugging support, which will show up in tight
> loops like this one (which we expect to be able to stop, for example,
> even when background tasks are disabled).

Yes, I've run into problems with real-time apps where debug sessions  
are too slow to be usable but release builds happily run over 60 FPS.  
It can be tough to properly debug things in that situation.

>> However the report
>> makes some incorrect assumptions as well - basically this test is
>> flawed. Daniel's code is making a function call in the dylib case and
>> no function calls in the RB case so if all things were equal RB
>> should've been _way_ faster, not just marginally faster.
>
> Again I wonder why you say .SingleValue isn't a function call.  Have I
> missed something?

Yup!

Frank.
<http://developer.chaoticbox.com/>


_______________________________________________
Unsubscribe or switch delivery mode:
<http://www.realsoftware.com/support/listmanager/>

Search the archives:
<http://support.realsoftware.com/listarchives/lists.html>

Reply via email to