Betis
![]() |
I think that above each component there should be a little yellow box that says how many seconds it takes to render it. This would be very useful for optimization, and finding out which parts of your filter aren't very optimized.
There would obviously be an option to turn these off, or have them only appear on hover, or something, that's FF's choice. But the main thing. ![]() ![]() (Click this for full size) Roses are #FF0000
Violets are #0000FF All my base are belong to you. |
|||||
Posted: October 5, 2008 5:17 pm | ||||||
Kraellin
![]() |
+1 on this.
If wishes were horses... there'd be a whole lot of horse crap to clean up!
Craig |
|||||
Posted: October 5, 2008 9:17 pm | ||||||
jffe |
We suggested that back in the beta days, seems like if I recall Vlad basically said to RTFW (W = wiki) or something along those lines, if you wanted to learn to make more efficient rendering filters haha
![]() jffe Filter Forger |
|||||
Posted: October 5, 2008 10:16 pm | ||||||
Betis
![]() |
I see what you're saying, and it is meant to be based off of what's plugged into it.
Besides, it is Optional, I would keep it on, even if it takes an extra second, it takes that long to get to different parts of the filter anyway. Roses are #FF0000
Violets are #0000FF All my base are belong to you. |
|||||
Posted: October 5, 2008 10:23 pm | ||||||
Kraellin
![]() |
you know, i still think this is a good idea. but you dont base it on the component; you base it on how long that component preview/thumb image took to render. you could even do it as a component where you simply inserted the 'Time' component into your algorithm anywhere you wanted and it would simply return a value. it would be a good way to debug a slow filter or speed up an average one.
If wishes were horses... there'd be a whole lot of horse crap to clean up!
Craig |
|||||
Posted: October 5, 2008 10:27 pm | ||||||
jffe |
It's not a bad idea, I just think Vlad already axed it way back, but it can't hurt to ask again especially with a mock-up screen pic of it.
![]() jffe Filter Forger |
|||||
Posted: October 5, 2008 11:19 pm | ||||||
Carl
![]() |
||||||
Posted: October 6, 2008 2:55 am | ||||||
CorvusCroax
![]() |
+1
Some kind of feedback system for render times would be very helpful, IMHO. As it is now, we're sort of groping in the dark as to the benefits of constructing one way vs another. Some kind of feedback on render times would help users learn better filter construction. Just some way to benchmark your filter... Even if it was a little 'tool' you ran inside the filter editor, which then output the time at each step (ie not always-on). Or even if was just a 'total weight' of the components. |
|||||
Posted: October 14, 2008 3:00 pm | ||||||
Betis
![]() |
Better filters mean less Low Usage Filters, More High Usages, and overall good appeal in the face of Filter Forge!
![]() Roses are #FF0000
Violets are #0000FF All my base are belong to you. |
|||||
Posted: October 14, 2008 5:32 pm | ||||||
Crapadilla
![]() |
Betis, we've discussed something like this here.
![]() --- Crapadilla says: "Damn you, stupid redundant feature requests!" ;) |
|||||
Posted: October 15, 2008 5:52 am | ||||||
Betis
![]() |
It seems you have! At first it was different than mine, samples and whatnot, but at the very end Craig posted something about seconds, and that I agree would be the same here.
The overall idea about speed reports about your components is a good one. I think Vlad should get in here and talk to us about it. Roses are #FF0000
Violets are #0000FF All my base are belong to you. |
|||||
Posted: October 15, 2008 7:31 am | ||||||
Vladimir Golovin
Administrator |
I'm watching but I'm not willing to talk at this point ![]() |
|||||
Posted: November 26, 2008 11:53 am | ||||||
Crapadilla
![]() |
Vlad, how about providing a separate compile that provides the render troubleshooting & analysis functionality we talked about here and in the other thread linked above?
This way, the average user would not have his renders slowed down by the additional 'analytical code', but a filter developer could - if necessary - launch the "troubleshooting" executable to get under the hood... ![]() --- Crapadilla says: "Damn you, stupid redundant feature requests!" ;) |
|||||
Posted: November 26, 2008 12:07 pm | ||||||
Kraellin
![]() |
sort of like having the 'render time' option we already have in the main gui? turn it on, turn it off, dilla? that could work.
If wishes were horses... there'd be a whole lot of horse crap to clean up!
Craig |
|||||
Posted: November 26, 2008 12:41 pm | ||||||
Betis
![]() |
I like Dilla's idea.
Maybe not a separate exe though, like Kraellin is saying.
And what point would you be thinking about joining? ![]() ![]() Roses are #FF0000
Violets are #0000FF All my base are belong to you. |
|||||
Posted: November 26, 2008 3:36 pm | ||||||
Crapadilla
![]() |
I was remembering something that Vlad posted a while ago regarding render diagnostics, but it turned out to be in another thread than the one I linked to above:
So, the idea was to provide a separate executable (say 'FilterForgeDiagnostics.exe Vs. FilterForge.exe) that features render diagnostics. According to Vladimir this won't be an option inside the regular FF compile since it would slow down things for everyone. --- Crapadilla says: "Damn you, stupid redundant feature requests!" ;) |
|||||
Posted: November 26, 2008 7:20 pm | ||||||
Betis
![]() |
Oh I see. I wouldn't be too opposed to having a separate exe anyway. A bit of an inconvenience, yes. But nothing too bad.
![]() Roses are #FF0000
Violets are #0000FF All my base are belong to you. |
|||||
Posted: November 26, 2008 7:44 pm | ||||||
Crapadilla
![]() |
I bet he's got something up his sleeve that he's going to spring on us at a most unsuspected time! ![]() ![]() --- Crapadilla says: "Damn you, stupid redundant feature requests!" ;) |
|||||
Posted: November 26, 2008 7:57 pm | ||||||
Betis
![]() |
I can only wait...
![]() ![]() ![]() Roses are #FF0000
Violets are #0000FF All my base are belong to you. |
|||||
Posted: November 26, 2008 8:35 pm | ||||||
SpaceRay
![]() |
This is a very good idea for optimization of filters, and to be able to test many different components and combinations and be able to see and test which one would be the fastest. But do not know if now 4 years after, Vladimir could have changed of idea and willing to talk about this or keep only watching ![]() |
|||||
Posted: February 16, 2012 11:50 am | ||||||
SpaceRay
![]() |
In February 2012 I have posted about this 2008 thread fr om Betis, but it was not the moment for Vladimir to talk about this
Now one year after, in February 2013 is already making the FF 4.0 and I want to bring back this topic that Betis started as I think it would be really helpful and useful to optimize and build the filters in a better way and get faster render results KNOWING wh ere there could be some time rendering problems between the filter components and links. This could perhaps help much to solve possible problems like the one Morgantao found in this thread --> FF math: 30+1=163 I have seen that the same as it happens here, it can happen also with others combinations of filters that are good separate and when joined in one filter, the render time is much higher than the sum of the two |
|||||
Posted: February 6, 2013 4:54 pm |
Filter Forge has a thriving, vibrant, knowledgeable user community. Feel free to join us and have fun!
33,711 Registered Users
+18 new in 30 days!
153,531 Posts
+39 new in 30 days!
15,347 Topics
+72 new in year!
355 unregistered users.