Hi Jonah, If you would like I’m happy to be paid By Autodesk to learn Bifrost in a public way!
I did that with ICE (although unpaid) and posted videos as I went, which became very popular and I’d consider myself an ICE expert now. People like watching other people learn as it’s more casual and less intimidating than watching an expert go through the motions. Paul Sent from my iPad > On 19 Feb 2020, at 17:57, Jonah Friedman <[email protected]> wrote: > > > Hi Chris, > > I was also an avid ICE user for many years, and following that Fabric Engine. > Being exposed to ICE was a turning point in my career in fact. > > I'm going to try to illustrate the difference with a simple graph that > contains very familiar pieces that I can use to highlight the difference.This > graph deforms a mesh with an f-curve on the x axis, applied to the Y > position. The green grid is before, the white is after. > > <image.png> > > > The graph is doing familiar operations to ICE users with familiar names: > get_point_position, converting the vector into scalars, applying an f-curve > to one, and rebuilding the vector and setting the point positions all have > essentially the same names and meanings in this graph. However the graph is > different. (Note this is an internal build, if you notice discrepancies in > the screenshot). > > To highlight the differences: > > 1. We can't just get point positions, we need to bring in a mesh first and > get positions from that. Maya does not have the operator stack that XSI had, > and so the graph is its own entity, and not a part of the mesh. Instead, the > mesh is just an input. > > 2. The part where we adjust the positions is operating on the per-point array > of positions. However in the graph, this is just flowing data. Those three > nodes in the center are processing the arrays. Ports shaped like hats instead > of squares signify arrays. The graph is not operating in the context of > per-point, but instead is just processing whole arrays in one shot. > > 3. "set point position" needs a geometry to set point positions on. We can't > set point positions on "this", we have to set them on data flowing in the > graph, and there is no "this". The mesh is flowing through the graph on the > purple connection. > > 4. In the end we output the shape, and it appears in Maya as a "bif shape". > This can be converted back to Maya geometry if you like, but is also > renderable if needed. You could set up a Bifrost graph as a more-or-less > traditional Maya deformer setup, but it'd require a bit more setup. (We're > improving this too with improvements to the Maya integration). > > The fact that this mesh is flowing through the graph just like any other kind > of data, buildable and modifiable, as pure data flow is the key difference > from ICE to understand. Whole geometries can be created in the graph in > multi-stage processes. Create_mesh_sphere outputs a sphere mesh and is > programmed in the mesh, and that sphere can be used downstream for any > purpose. > > --- > > Visualizing data: Watchpoints are not the same as the similar feature in ICE, > the debug drawing. Instead right now to do visual diagnostics, you can use > the "scope" nodes whose job it is to visualize data. For example, see the > image below of "volume_scope" which visualizes volumes. This is working by > generating strands in the graph to visualize the tile tree boxes (left) and > flowlines in the volume (right). We're working on improving this as well. > > <image.png> > > > To see the data in whole in numeric / text form, "dump_object" is your best > bet which sends the data to a text file. In there you can see all the data > that defines the geometries except volumes, which are still opaque for now. > > I'd also like to say it's a major concern of mine that Bifrost is not the > easiest to learn and understand. It's not something we have a silver bullet > for - some of the great things that ICE did are tricky for us and visual > programming is not the easiest thing even in the case of ICE. We do however > have multiple efforts underway to improve the situation. One nice example > that's in right now that you may or may not have noticed - most nodes we ship > with are documented in-app, in an info tab. > > <image.png> > > > Some other info: > > Docs: > https://u9432639.ct.sendgrid.net/ls/click?upn=A5uD99yDGgJFqsHo0L78rmOwSniDKlMnd1Pg9yA-2BXHyaQxEyE-2B-2Bc87uWT-2B7J28U-2BKD66UDPyRLB5thkwKYg-2FjA-3D-3D5hTF_x9fWPgxQbfi69QJnHJqUKZsAJHrwlN1lgOIh62WX2fRJmJjuULg1gUn5oOsOK5zaq0VJCxCMthYbtptP0xmApMcWkXJiscj3WuOzXMt18PNieAXUgKv1boHspus0mJOhCdqluGM587WEIJ148nUpL-2F4b87S1Q4boLcCqwDPuwQSBwCwUwoJyiHIE-2Bxfkf0yQIv-2BQ8iGfofUkZMQYKp1JAq-2B7qG2vWIJcdHgCU0q3jUx-2FGo8rlNh235HXKptpprAY > Release notes from our last release, released yesterday: > https://u9432639.ct.sendgrid.net/ls/click?upn=A5uD99yDGgJFqsHo0L78rmOwSniDKlMnd1Pg9yA-2BXHyaQxEyE-2B-2Bc87uWT-2B7J28U-2Bxc1KOAa00-2FVd-2B1Yw39zW6-2BFbjO0hxPee6psQRPM3SuS7dm-2FzLXGqbZOwO6cYdY3qZ78buEf5QUpnPvzDa2duQI5w7V9iT-2F9FwwfYCBAiOlU-3DfQXw_x9fWPgxQbfi69QJnHJqUKZsAJHrwlN1lgOIh62WX2fRJmJjuULg1gUn5oOsOK5zaq0VJCxCMthYbtptP0xmApI0uFK1akNE9g1EtdBtkKmJ-2BN-2BWf7DaQjhywAl1EEsW0SB9oHWK5MjsP-2FMKyv7KKZB9v4RHl-2BSEoBl80oUtU-2FCBSBLILFdGxBOTgPEC-2BZbwJXdo-2BAAx2KhXxBwuikKSj43C0rPhqYHZA0n7GViQFdrKhmSO95eYCjH9-2B-2FV4GIFs-2F > > Tutorials by Matt Chan: > https://u9432639.ct.sendgrid.net/ls/click?upn=A5uD99yDGgJFqsHo0L78rsOelCL9r8TPNhamPpv5mE1y9Smy6IOxvsFt-2FN5f361oaVyYm9tVo7gd7B-2FRXzKgPgbZYHOXq-2FG7I-2F9ss5bFlyc-3DioqK_x9fWPgxQbfi69QJnHJqUKZsAJHrwlN1lgOIh62WX2fRJmJjuULg1gUn5oOsOK5zaq0VJCxCMthYbtptP0xmApLtBEwy-2B8OseRSlDIkViH4bDaJVbLYvo8l7nTXBt4FX-2Bf-2FZJGyuJwIaPUzHW8k038dK-2FIyZrqk01kxUheRFBuZyZHILXk3DOjatpu60HmysVtWd532WKY01eKx1BwzVkr2vSKjh4SvBcgbdNk8uykw-2BrS1ZT5s-2FcDODBC8qXgTrX > > > >> On Tue, Feb 18, 2020 at 3:29 PM Chris Marshall <[email protected]> >> wrote: >> Oh Hi Jonah, >> Well ok as I’ve been quite a fan of ice, and it took me a while to warm to >> it but I can totally see it’s logic, all I’m seeing so far with Bifrost >> graph are limitations, compared to Ice. So putting all that to one side, I’m >> happy to start looking at it again afresh. What would help is a document >> that describes, with simple examples, what the structural differences are >> between the two systems and how bifrost graph is so open and powerful. ICE >> was / is a very comprehensive and what I’d describe as a mega flexible >> system, once you know what you’re doing. Some would probably argue it was >> too flexible for artists as it required some coding logic to really push it. >> To be honest, with Bifrost graph I’ve dipped in a few times and got nowhere. >> A basic fundamental description of how the system is supposed to used would >> help, and why this might be different to ICE. >> >> Thanks >> >> Chris >> >> >> >> >>> On Tuesday, 18 February 2020, Jonah Friedman <[email protected]> wrote: >>> Hi Chris, I'm the Product Owner of the Bifrost graph. I'd be happy to >>> answer any questions about it. What specifically are you trying to do? >>> >>> The graph is different from ICE in several ways, but it's very open and >>> very powerful. >>> >>>> On Tue, Feb 18, 2020 at 4:16 AM Chris Marshall <[email protected]> >>>> wrote: >>>> OK So is Bifrost graph years away from getting close to ICE or am I >>>> missing something? >>>> I really am trying to figure it out, but either I'm doing something wrong >>>> or it's very very limited. >>>> Anyone got anywhere with it? >>>> Thanks >>>> Chris >>>> >>>> ------ >>>> Softimage Mailing List. >>>> To unsubscribe, send a mail to [email protected] >>>> with "unsubscribe" in the subject, and reply to confirm. >> >> >> -- >> Chris Marshall >> Mint Motion Limited >> 029 20 37 27 57 >> 07730 533 115 >> https://u9432639.ct.sendgrid.net/ls/click?upn=A5uD99yDGgJFqsHo0L78rjo3fW-2BI05z8hPddj-2BYSrjut8V8Kr3SyUl8pLj-2FCGT-2FDPGdYFuSQLHhkvggY1PfxbZQfR-2FKexA2KHRyXTiroKSozHjiRnNNMUP0xHbNxyhdpl7aNCnZETX869nBLoCNE34ls3m75bkN1ArPnFW7nOrLX3T8LZlP4gIKSKMrOxUj2vmUcd5vgX-2Bi6Pmvm4YoKBKyvkOGk7sSKUsN9kwKYhSvgZl3kmU2u5rCSwHSwrzdOGCJNjsECcPFlGnOVZFnLa8FzFtdi4srPxLg531GfxROSfbwXquUzPNfuHftg9-2FduO23CMSY-2BbvHOoKtkRkLp5rHdszec9P1YXy-2FUvr-2FX0HYgG0z-2B0NWUmyPhISEznWtltyoB_x9fWPgxQbfi69QJnHJqUKZsAJHrwlN1lgOIh62WX2fRJmJjuULg1gUn5oOsOK5zaq0VJCxCMthYbtptP0xmApHAK89UvqB2kwRlUDGDz7TbdT7mBweDOQmpdaey345qedYLAhVS9nzq3-2F9tJMqYPrBGCVKXgLstouoPGX8bK80UApy4Z2qR1YulG71Sejk3K3qltHBGSPbVPhX4DLimxvtQDI4Io5hf3sRFzfPmY8M9QvHg4GH228wb8E2wkckM0 >> >> https://u9432639.ct.sendgrid.net/ls/click?upn=A5uD99yDGgJFqsHo0L78rjo3fW-2BI05z8hPddj-2BYSrjut8V8Kr3SyUl8pLj-2FCGT-2FDPGdYFuSQLHhkvggY1Pfxbf3Pjd3v7OjhHXmQizfgqwnfqyP01q1NPy7V0BC8TKQJiWmZMfbqWyM4npcaC3dXgRrbIEePuHTx517jbRZUCa641TfW9NIIREgksqe6LPeEW-2BOizK670EUc993DA9tI7mCh2QvOZkaGIggq5zX8e7OVE-2Ff9WEWApk9YrvkskoI1hBXR9ZmqHwwjmHEZPBG1AEQOAWH4ZhlfYJlBEfAtyLAE9UEvypwj-2FosRlByAc-2FIrIBjuJ6IhUluh2e22-2B9pS63tB3yNHwmce3t9wKXPV-2Bak-3D4YM5_x9fWPgxQbfi69QJnHJqUKZsAJHrwlN1lgOIh62WX2fRJmJjuULg1gUn5oOsOK5zaq0VJCxCMthYbtptP0xmApDqjtS-2F-2BI-2BkTcfIumk1Lngb-2FIvV8AOfrQGV4XGN817t5gS-2FpNU5SgHCLotDi4Ct0saD4uOjfa3wVwCiX8cMbLgCp4mJMCcHzQKkFRYNfTYnC9GbkuZbgG65dURjyGKYCLN0EoeOXaf3oLC5otswIIR4MDTbw10im1NQ8mJEVpXBc >> >> >> >> >> ------ >> Softimage Mailing List. >> To unsubscribe, send a mail to [email protected] with >> "unsubscribe" in the subject, and reply to confirm. > ------ > Softimage Mailing List. > To unsubscribe, send a mail to [email protected] with > "unsubscribe" in the subject, and reply to confirm.
------ Softimage Mailing List. To unsubscribe, send a mail to [email protected] with "unsubscribe" in the subject, and reply to confirm.
