Well I wasnt meaning to imply that I could upgrade hardware to get round these performance issues, I will be upgrading my GPU only once Apple 'reattach the limb'.
I dont think its really possible to talk about the future here properly due to NDAs but I will say that I am more optimistic than George seems to be. Cheers Steve On 22 Mar 2010, at 21:15, George Toledo wrote: > It may be an interesting tell to disable the iterator macro and see what > performance gains you achieve. > > In regards to Steven Watkins comment, a powerful GPU is like a bandaid on a > severed limb, the limb being vital QC patches which have regressed in > performance do to extremely poor conceptualization, and new ones that never > worked correctly to begin with, and are currently more broken than they were > in the GM ( Open CL, Mesh Warps, Mesh Lines, etc.) What would be great is if > they could be immediately regressed in an update (especially the iterator), > but I think it's more likely to not be so immediate. > > That said, I have not found many situations where there are not workarounds > in 10.6.x (although they are sometimes less obvious, the same visual result > can be achieved many ways). > > Best regards, > George Toledo > > On Mon, Mar 22, 2010 at 5:04 PM, Steven Watkins <[email protected]> wrote: > Yes I would be inclined to blame 10.6.2, which I generally think of as a > disaster when it comes to QC. Its almost surprising that there hasnt been > more ranting about it on the net. > > I doubt we have too much longer to wait till this sad moment in the evolution > of QC passes. I look forward to QC on Snow Leopard getting back on track, > assuming they sort out some of the problems I will be getting a powerful GPU > for my Mac Pro with a view to getting OpenCL to do some interesting things. > > I think your macbook pro only has the 9400M graphics card but that should > still deliver good enough performance in many areas, I doubt it is the > problem with performance of this older composition. > > Cheers > > Steve > > On 22 Mar 2010, at 19:30, Tamas Nagy wrote: > > > Are you using 10.6.2? There are some performance bug with QC on 10.6.2 as i > > know. > > > > > > On Mar 22, 2010, at 7:35 PM, arri wrote: > > > >> Hi, > >> > >> i didn't find any other posts on similair issues and i would actually > >> have been surprised > >> if i had. because i think there's really something wrong: > >> > >> today i opened a fairly simple patch: "Constructivism" by Futurismo > >> Zugakousaka. (a favorit) > >> http://ibeatsonline.com/index.php%3Fref=project-quartz-jp-quicktime-26_Constructivism.html > >> > >> it had been quite a while since i last worked with QC, and it was the > >> first time to work with QC on > >> my new 13" macbook pro (latest model). having high expectations, it > >> was a 'bit dissappointing' > >> to see this patch would not even run at 15 fps (!!!) while the cpu is > >> at max, as if all hardware- > >> rendering is disabled (which it is not) > >> > >> i often work in a similair graphical flat/2d style myself, also using > >> text, letters/typography alot and > >> most of the time this is a joke for video hardware to render. so i'm > >> extra curious why this performs > >> so bad. it's is not a high-end machine, but it's new and also it's not > >> a complex patch > >> > >> i tested it on my old '06 mini (1.66Ghz-coreduo, intel-GMA950) and no > >> surprise: steady 60fps > >> no problem and the CPU picking its' nose.. > >> > >> > >> i think that either there's something really wrong with my macbook, or > >> maybe i'm just not informed > >> on modern hardware, and should learn about howto build patches conform > >> todays standards.. > >> > >> or is it yet something else? > >> > >> > >> also, i have -some- exprience with OpenGL and quite alot with Quartz, > >> so i thought i might be able to > >> use some of that knowledge to learn more about the issue. i didn't > >> think QuartzDebug would make > >> miuch sense here so i looked at "OpenGL Profiler" and "OpenGL Driver > >> Monitor" . > >> both seem to generate a huge load of interesting data, but i haven't > >> really got an idea (yet..) of howto > >> interpret this data in a sensible way. before i dive into that, does > >> this approach make any sense? > >> > >> > >> > >> > >> > >> thanks > >> arri > >> _______________________________________________ > >> Do not post admin requests to the list. They will be ignored. > >> Quartzcomposer-dev mailing list ([email protected]) > >> Help/Unsubscribe/Update your Subscription: > >> http://lists.apple.com/mailman/options/quartzcomposer-dev/tamas.lov.nagy%40gmail.com > >> > >> This email sent to [email protected] > > > > _______________________________________________ > > Do not post admin requests to the list. They will be ignored. > > Quartzcomposer-dev mailing list ([email protected]) > > Help/Unsubscribe/Update your Subscription: > > http://lists.apple.com/mailman/options/quartzcomposer-dev/steve%40dvmachine.com > > > > This email sent to [email protected] > > _______________________________________________ > Do not post admin requests to the list. They will be ignored. > Quartzcomposer-dev mailing list ([email protected]) > Help/Unsubscribe/Update your Subscription: > http://lists.apple.com/mailman/options/quartzcomposer-dev/gtoledo3%40gmail.com > > This email sent to [email protected] >
_______________________________________________ Do not post admin requests to the list. They will be ignored. Quartzcomposer-dev mailing list ([email protected]) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/quartzcomposer-dev/archive%40mail-archive.com This email sent to [email protected]
