[Tremor] LOW_ACCURACY mode and MDCT improvements

[Mike Giacomelli]

> On 03/25/2010 07:08 PM, Mike Giacomelli wrote:
>> Then test without sending it to disk. Or measure how long it takes to send
>> to disk and subtract that. I don't know, it seems like this is not a difficult
>> problem.
>
> i already did that. that's the tests i've been reporting to this list
> over the last few days. sorry if that wasn't clear the first time
> through. the only difference between what you wrote and what i had
> already presented was sending the data to disk, which was why i was
> explaining my reluctance about that part.
>


Great!


I saw before you posted results for the patch with ASM.  Would you mind
reverting the patch (but leaving in ASM) and benchmarking it again so we
can compute the percentage speed up?  Thats really what I'm curious about. 		 	   		  
_________________________________________________________________
The New Busy is not the old busy. Search, chat and e-mail from your inbox.
http://www.windowslive.com/campaign/thenewbusy?ocid=PID27925::T:WLMTAGL:ON:WL:en-US:WM_HMP:032010_3