YOUR ACCOUNT

Sphinx.
Filter Optimizer

Posts: 1750
Filters: 39
It would be useful with a simple boolean property to bypass the bitmap cache and pass through source samples in conventional procedural manner.

The reason for this is easier empirical testing of performance improvements and/or qualitative impact. Also this could be used to give the user the final decision about quality vs speed.
  Details E-Mail
GMM
Moderator
Filter Forge, Inc
Posts: 3491
We plan to implement this functionality via a separate Bypass component. I'm afraid I can't give an ETA for its appearance.
  Details E-Mail
Vladimir Golovin
Administrator
Posts: 3446
Filters: 55
Quote
GMM wrote:
separate Bypass component


Not a component, of course. We'd like to implement a proper Bypass feature, but this is not a trivial thing to do, because a proper bypass would require two internal component trees -- one visible to the user in the Editor, and another one for the renderer with all the bypassed components thrown out. This is a significant rewrite because the current Editor uses the same single tree for both the UI and the renderer.
  Details E-Mail
Sphinx.
Filter Optimizer

Posts: 1750
Filters: 39
A real infrastructural Bypass would be awesome and preferable. I understand why that is alot work smile;-)

A simple internal bypass (i.e. f(x) -> x) would also do - perhaps as a first edition before the "significant rewrite".

Anyway, I'm glad to hear you're considering the feature smile:-)
  Details E-Mail
Vladimir Golovin
Administrator
Posts: 3446
Filters: 55
Quote
Sphinx. wrote:
Anyway, I'm glad to hear you're considering the feature


We even have a preliminary design for it:

  Details E-Mail
Vladimir Golovin
Administrator
Posts: 3446
Filters: 55
Here's a more recent pic that looks a bit better:

  Details E-Mail
Sphinx.
Filter Optimizer

Posts: 1750
Filters: 39
Looks good smile:-)

How is it going with the work on "The Next Big Thing"? Need testers?
  Details E-Mail
Vladimir Golovin
Administrator
Posts: 3446
Filters: 55
Quote
Sphinx. wrote:
How is it going with the work on "The Next Big Thing"? Need testers?


In short, GPUs are weird, hard to program, and we still have no idea what to do with them on Mac, especially given Apple's transition to their own silicon.

And BTW, the screens aren't from The Next Big Thingy -- they are from an Editor overhaul we're planning for normal, regular Filter Forge.
  Details E-Mail
emme
Posts: 718
Filters: 8
So, is The Next Big Thingy currently on pause, or still in development? Too bad about the GPU difficulties, but it's understandable.

The new FF editor designs look clean. Looks like there might be some new functionality along with the new visuals? Is this planned for version 10 or for later?
  Details E-Mail
Vladimir Golovin
Administrator
Posts: 3446
Filters: 55
Quote
emme wrote:
So, is The Next Big Thingy currently on pause, or still in development? Too bad about the GPU difficulties, but it's understandable.


No, not on pause. The current focus is a tile-based / render-on-demand architecture for GPUs, which is needed for rendering anything bitmap-based.

Quote
emme wrote:
Looks like there might be some new functionality along with the new visuals?


There'll be a lot of new functionality -- dragging connectors from inputs (not just from outputs as before), click-click-connections in addition to dragging, ability to reconnect connections coming from an output, component creation menu available during connecting (like in Houdini / Unity / Unreal), dots for rerouting connections, named pins for distributing connections, named wireless pins for laying connections through the air, ability to rename components, collapsible component previews, framed groups (like in Unity), and a proper bypass -- but this one may take a while.

Oh, and design-wise, every single thing is now aligned to the grid, which hopefully will somewhat mitigate spaghetti networks facilitated by the current editor.

Quote
emme wrote:
Is this planned for version 10 or for later?


That would be nice, but the development may take longer. We'll see.
  Details E-Mail
Sphinx.
Filter Optimizer

Posts: 1750
Filters: 39
This is indeed exciting news smile:-) Looking forward to that update.
  Details E-Mail
emme
Posts: 718
Filters: 8
Awesome! This will really make the workflow more robust.

One small but useful feature would be the ability to duplicate connections fr om inputs with a modifier key (like ctrl). There's a connection from A -> B and I want to connect A -> C, I could simply ctrl + drag from the B input to C input and duplicate the connection. Useful for large trees wh ere the components can be very far from each other.

Maybe the features you listed already solve this smile:)
  Details E-Mail
Vladimir Golovin
Administrator
Posts: 3446
Filters: 55
Quote
emme wrote:
Maybe the features you listed already solve this


No, they don't smile:) This sounds like a good idea.
  Details E-Mail

Join Our Community!

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!

153,533 Posts
+38 new in 30 days!

15,348 Topics
+73 new in year!

Create an Account

Online Users Last minute:

34 unregistered users.

Recent Forum Posts: