It's not that they're necessarily faster, it's that you get more manual control and may be better if you're happy to trade off the automagic cleaning of cache data, higher precision vs half float exr, and not needing to explicitly render you get with diskcache.
If you've got 'check file matches input' on it'll need to calculate the hash of the tree above to check it (see rest of article) - cheaper than figuring out the image but could still be doing work, so that might be what you're seeing. See if switching it off/using a non-hash-metadata writing format like dpx makes any difference, as perhaps that's it. Might explain why different script configurations see different times spent calculating. J On 8 July 2015 at 10:48, Charles Bedwell <[email protected]> wrote: > That's quite interesting, thanks for posting it. > > The author mentions that using read from write nodes as being a faster > and perhaps more efficient way of not recalculating upstream nodes than > using the diskcache node. > > The issue I have is that it is not faster at all, upstream is still > being calculated and the read from write is completely ignored, and I'm > stuffed if I know why. If I make a dedicated read node and read that > sequence, connect it to the downstream it's fine. Plays back like > lightning. Putting a readfromwrite node in the stream does not seem to > function. > > Charlie > > On 8 Jul 2015, at 10:21 am, Jack Binks <[email protected]> wrote: > > CAUTION EXTERNAL EMAIL > > > > > This may be informative - there's a section on diskcache in particular, > but the rest is good to know if you haven't come across it before: > > http://major-kong.blogspot.co.uk/2012/11/all-your-cache-are-belong-to-us.html > > J > > On 7 July 2015 at 15:04, Charles Bedwell <[email protected]> > wrote: > >> Hi Daniel, >> >> At the moment Nuke 9.05. >> >> Elias, I rendered out, checked Read and pointed the viewer to that (exr >> render, check file matches input disabled) so the write/read is sitting at >> the bottom of the tree. Playback still awful (I can see the calculating >> keyframes or whatever it is window pop up briefly from a tracker above it). >> >> The only thing that worked is to disconnect the tree and make a read >> node pointing to the render. Playback is fine as it's just playing back the >> exr seq. >> >> >> >> >> On 7 Jul 2015, at 2:54 pm, Daniel Hartlehnert <[email protected]> wrote: >> >> CAUTION EXTERNAL EMAIL >> >> >> >> >> Hi, >> >> i would be curious about this too. Because i expierenced the same >> behaviour in many Nuke versions and finally just gave in, thinking it just >> doesn't work. >> I'd be happy to hear other feedback though. >> What Nuke version are you using? >> >> Cheers, >> Daniel >> >> Am 07.07.2015 um 15:43 schrieb Charles Bedwell: >> >> Hi, >> >> Is there some secret to using the DiskCache node? I have a large node >> tree which I'm pretty much done with and would like to cache it so I don't >> have to render it during playback as it's rather slow. >> >> I added a DiskCache node and clicked PreCache, but when I scrub my >> timeline the nodes above it flash yellow as if they are processing. I >> haven't changed the zoom level or otherwise changed anything. This is with >> the viewer looking at the diskcache node. >> >> Am I doing something wrong? >> >> >> >> >> *CHARLES BEDWELL* >> Compositor, Encore London >> 142 Wardour Street, London, W1F 8DD >> t: +44(0)20 7149 2409 <+44(0)20%207149%202409>/ m: +44(0)7789 190206 >> <+44(0)7789%20190206> >> www.encorepost.co.uk| FACEBOOK <https://www.facebook.com/encorepost> | >> TWITTER <https://twitter.com/EncorePost> >> >> >> <115070714433600569.jpg> >> P Please consider the environment before printing this email. >> _______________________________________________ >> Nuke-users mailing list >> [email protected], http://forums.thefoundry.co.uk/ >> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users >> >> >> DO NOT open attachments or click on links from unknown senders or >> unexpected emails >> >> >> >> >> >> This e-mail and any attachments are intended only for use by the >> addressee(s) named herein and may contain confidential information. If you >> are not the intended recipient of this e-mail, you are hereby notified any >> dissemination, distribution or copying of this email and any attachments is >> strictly prohibited. If you receive this email in error, please immediately >> notify the sender by return email and permanently delete the original, any >> copy and any printout thereof. The integrity and security of e-mail cannot >> be guaranteed. >> >> _______________________________________________ >> Nuke-users mailing list >> [email protected], http://forums.thefoundry.co.uk/ >> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users >> >> >> _______________________________________________ >> Nuke-users mailing list >> [email protected], http://forums.thefoundry.co.uk/ >> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users >> > > DO NOT open attachments or click on links from unknown senders or > unexpected emails > > > > > > This e-mail and any attachments are intended only for use by the > addressee(s) named herein and may contain confidential information. If you > are not the intended recipient of this e-mail, you are hereby notified any > dissemination, distribution or copying of this email and any attachments is > strictly prohibited. If you receive this email in error, please immediately > notify the sender by return email and permanently delete the original, any > copy and any printout thereof. The integrity and security of e-mail cannot > be guaranteed. > > _______________________________________________ > Nuke-users mailing list > [email protected], http://forums.thefoundry.co.uk/ > http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users > > > _______________________________________________ > Nuke-users mailing list > [email protected], http://forums.thefoundry.co.uk/ > http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users >
_______________________________________________ Nuke-users mailing list [email protected], http://forums.thefoundry.co.uk/ http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-users
