Re: [PanoToolsNG] Processor Speedtest
- Fourth System:
Hugin 0.7 beta5/nona: 0:12 min
Autopano Pro 1.4.0 alpha3: 2:16 min (i'm pretty sure there's a bug in the render engine for multi-core systems)
Operating System: SuSE Linux Enterprise 10
System Manufacturer: HP
Model: DL360 G5
CPU: Dual Quadcore Intel Xeon E5345 @ 2.33GHz
MemTotal: 20553984 kB
MemFree: 19725844 kB
SwapTotal: 7052464 kB
SwapFree: 7052464 kB
HDD: 2x10000 RPM SAS SFF Drives (RAID 1) with integrated Controller
- --- In PanoToolsNG@yahoogroups.com, Pat Swovelin <Panoramas@...> wrote:
>I do not have the Pro version, only tested PTGui 7.2
> On 8/4/2007 6:34 AM, Hans Nyberg rambled on about ...:
> > --- In PanoToolsNG@yahoogroups.com, "erik leeman" <erik.leeman@> wrote:
> >> I did some more runs to see if the results were reliable.
> >> Not very I'd say now. The spread is just to wide between runs with
> >> identical settings : (
> > Well I just discovered what seems to be a bug in either panotools or
> > PTGui Windows.
> > PTGui 7.2 Panotools from Joost and the fast pano12.dll from Jim Watters
> > Try this.
> > 1. Stitch using PTgui but warp with panotools. no fast transform
> > 2. Just book the use fast transform and stitch again without
> > restarting PTgui
> > 3. unbook fast transform and stitch without restarting PTGui
> > 4. Save the last settings and close. open again and stitch.
> > You will see that the fast transform reduces time with around 30% but
> > it does not change
> > back to No fast transform before you save it and restart,
> > On Mac the fast transform does not have any effect at all with
> > panotools warping, only
> > with PTgui
> Was this with PTGui 7.2 or PTGui Pro 7.2 (there may be a difference)?