PanoTools mailing list archive

Mailinglist:PanoTools NG
Sender:Bernhard Vogl
Date/Time:2008-Jan-04 00:33:13
Subject:Re: Re: New hardware testing on PTgui performance

Thread:


PanoTools NG: Re: Re: New hardware testing on PTgui performance Bernhard Vogl 2008-Jan-04 00:33:13
Hello Joost,
>>> As has been mentioned in this thread, IO related bottlenecks
>>> are due to the use of uncompressed fileformats, if not for
>>> the final then for intermediate tempfiles. A request to 
>>> the authors of frontends to my pano-libraries (ptgui, nona,...): 
>>> please insert the line
>>>
>>> TIFFSetField(tif, TIFFTAG_COMPRESSION, COMPRESSION_PACKBITS );
>>>       
>
> Actually PTGui doesn't use tiff files for temporary storage.. 
> Compression could be incorporated somehow, but it's not that simple.
>   
The more i produce some repeatable results, the more i have to retreat 
my opinion that temp file size is the most important issue during 
stitching of large panoramas. In my worst cases, warping (and temp file 
writing) consume not more than 1/4 of the total stitching time.
This - of course - is only true as long as you stay with PTGui's 
internal warping and blending. Using an external blender will expose 
other timings...
>   
>> Just as a number: I have compared 2 methods with PTGui:
>> - one is rendering a large panorama as one large image: took 2h 38min, 
>> warping (and temp file writing) took only 30min, the rest was blending time
>> - the other is rendering this large panorama in 100 parts (slices): took 
>> 1h 25min
>> It can easily be seen that computational cost of the blender raises 
>> exponentially if you overstrain current blending algorithms with large 
>> images.
>>     
>
> That's interesting indeed. I have several ideas to improve performance 
> and reduce disk usage for gigapixel panos but haven't yet found the time 
> to do this.
>   
That sounds great!
The better the hardware, the more blending will be the issue. I have 
added another machine to the test chart that shows the problem pretty 
clear: 8 cores, 32GB RAM, 4 10k 2.5" SAS disks in RAID 0 mode as temp 
partition: warping takes 7 minutes but total stitching time is still 1h 
42min....this is only 1.55 times faster than a ordinary desktop...

Best regards
Bernhard


-- 
<*> Wiki: http://wiki.panotools.org
<*> User Guidelines: http://wiki.panotools.org/User_Guidelines
<*> Nabble (Web) http://www.nabble.com/PanoToolsNG-f15658.html
<*> NG Member Map http://www.panomaps.com/ng
<*> Moderators/List Admins: #removed# 
 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/PanoToolsNG/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/PanoToolsNG/join
    (Yahoo! ID required)

<*> To change settings via email:
    mailto:#removed# 
    mailto:#removed#

<*> To unsubscribe from this group, send an email to:
    #removed#

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/
 

Next thread:

Previous thread:

back to search page