[gmpi] Re: Req issue: event timestamps
- From: "Jeff McClintock" <jeffmcc@xxxxxxxxxx>
- To: <gmpi@xxxxxxxxxxxxx>
- Date: Mon, 17 Nov 2003 08:10:56 +1300
> I'm waiting for someone to point out that the test is flawed because of
> something.. :)
Well, yes, you have shown that 64bit math is slower than 32 bit math.
You have not shown that plugins with 64bit timestamps will run 250% slower.
The 2nd half of the equation is "what % CPU does event processing use
anyhow?"
e.g. say event-processing code represented 1% my reverb plugin's CPU, which
uses 5% CPU overall.
in that case the difference between 64bit and 32bit events would be 0.05%.
In a softsynth, an event (note-on etc) may only occur once every 1000
samples. Event-processing just dosn't register on the CPU meter.
Of course, I am not fanatical about this. Offsets are 'do-able'.
But I am reminded about the 'old days', where we attempted to squeeze
every last byte out of our code. Later we all say "wasn't it dumb limiting
MIDI to 7bit values?, we need a new standard".
Best Regards,
Jeff
----- Original Message -----
From: "Tim Hockin" <thockin@xxxxxxxxxx>
To: <gmpi@xxxxxxxxxxxxx>
Sent: Sunday, November 16, 2003 5:32 PM
Subject: [gmpi] Re: Req issue: event timestamps
> On Sat, Nov 15, 2003 at 08:51:47PM -0700, Mike Berry wrote:
> > I ran this test on Windows XP (.NET 2003) on a P4 2.8 HT. My results
> > are:
> >
> > int32: 100%
> > int64: 188%
> > int64_2: 250%
> >
> > for what its worth.
>
> ouch. Admittedly, the cost of the DSP (which I left out) will dwarf the
> cost of the timestamps, but we should be frugal.
>
> I'm waiting for someone to point out that the test is flawed because of
> something.. :)
>
>
> ----------------------------------------------------------------------
> Generalized Music Plugin Interface (GMPI) public discussion list
> Participation in this list is contingent upon your abiding by the
> following rules: Please stay on topic. You are responsible for your own
> words. Please respect your fellow subscribers. Please do not
> redistribute anyone else's words without their permission.
>
> Archive: //www.freelists.org/archives/gmpi
> Email gmpi-request@xxxxxxxxxxxxx w/ subject "unsubscribe" to unsubscribe
>
>
>
----------------------------------------------------------------------
Generalized Music Plugin Interface (GMPI) public discussion list
Participation in this list is contingent upon your abiding by the
following rules: Please stay on topic. You are responsible for your own
words. Please respect your fellow subscribers. Please do not
redistribute anyone else's words without their permission.
Archive: //www.freelists.org/archives/gmpi
Email gmpi-request@xxxxxxxxxxxxx w/ subject "unsubscribe" to unsubscribe
Other related posts:
