[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: vpowf parameters backwards on Intel?

Subject: Re: vpowf parameters backwards on Intel?
From: Shaun Wexler <email@hidden>
Date: Mon, 24 Apr 2006 15:27:09 -0700
Delivered-to: email@hidden
Delivered-to: email@hidden

On Apr 24, 2006, at 1:50 PM, Ian Ollmann wrote:

This is a known problem that we intend to fix. <rdar://problem/ 4485735>

In certain cases, you might be able to use the vImageGamma functions or vForce instead. Apart from that, I'm not sure that there are any very good workarounds at the moment.

Not that it will help in the SSE case, but in AltiVec, a quick vpowf approximation is simply:

vFloat vPow = vec_expte( vec_madd( vExp, vec_loge( vBase ), (vFloat)(-0.0f) ) );

--
Shaun Wexler
MacFOH
http://www.macfoh.com

"Everywhere is within walking distance, if you have the time." - Steven Wright

_______________________________________________
Do not post admin requests to the list. They will be ignored.
PerfOptimization-dev mailing list      (email@hidden)
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/perfoptimization-dev/email@hidden

This email sent to email@hidden

References:
	>vpowf parameters backwards on Intel? (From: glenn andreas <email@hidden>)
	>Re: vpowf parameters backwards on Intel? (From: Ian Ollmann <email@hidden>)

Prev by Date: Re: vpowf parameters backwards on Intel?
Next by Date: Re: vpowf parameters backwards on Intel?
Previous by thread: Re: vpowf parameters backwards on Intel?
Next by thread: vectorOps.h
Index(es):
- Date
- Thread

Home