Re: OT:houdini stamp UV
Thanks, Yeah it works.. kinda. It gave me a very white pic, but the alpha was usable G *still loving houdini* On 10/06/2015 10:40, Cristobal Infante wrote: Right click on the node and select, Save texture UV to image. Never used it myself, let us know how it goes! Cris On 10 June 2015 at 09:21, Gerbrand Nel nagv...@gmail.com mailto:nagv...@gmail.com wrote: Hey guys. My dumb ass is asking dumb questions again. I cant find something similar to stamp uv in houdini Is it there, and how and where? Thanks G
Re: Virtual Apps
if you mean using a thin client on the desk, to connect to a remote workstation (in the server room) – then yes – have used this at a former studio. overall it worked quite well. on the thin client you would launch an app, on which you chose the workstation to login to and then a full screen window opens on which you see the workstations’ desktop – and you work you session. It’s very intuitive – apart from a few keyboard combos (ctrl-alt-del is on the thin client, so there’s a different combination to send that to the workstation) You could use the thin client at any desk to log in to any equipped workstation – handy at times – chaotic when your team members end up all over the place. The overhead on the workstation is pretty much zero. The added card handles the compression/communication – so you can push the workstation exactly as before. there was hardware compression/decompression of all signals – so it meant adding a dedicated card in the workstation - all data (kb, mouse, usb as well as monitors) goes through network. afaik the screen refresh is done on the thin client – which reduces the amount of data to be sent (no screens full of pixels) but also makes sure that despite long cable length, image quality is high . (compared to all KVM extenders I ever saw) To the very demanding artist there is a barely noticeable lag and some degradation – you can kind of make out the compression – but you do have to look for it. We decided on using the thin clients only for 3D artists, not for compositors. It would work for compositing most of the time, but when checking final images/shots, occasional little flicks or spots from the compression are disturbing. If you are the person who has 3 oversized monitors on his desk, and expects to have film quality visuals while modeling – this might not be for you. image quality can suffer from network load – as compression adapts some – and at a few peak moments network was so taxed (not because of the thin clients) that connections between clients and stations were lost massively. That’s unfortunate and real disruptive – but once the load was balanced again you would just login and the workstation was right where you left off – preferable to crashes and shutdowns. But it’s something to be aware of - if you have a problematic network, thin clients will add to the frustration. An added benefit was that there was much less heat generated and electricity used in the office rooms – in small cramped, badly ventilated and badly equipped offices that can be a tangible benefit. I have memories of humming workstations under desks, burning desklights and running ventilators everywhere (including on an opened workstation case which is a very bad idea) creating an unpleasant and unhealthy microclimate. The switch to thin clients was heavenly. As were LED desklights. Hope it helps some. It’s a big step – that you need to consider carefully with your supplier (ours was HP) – and ideally in a riskfree way, where you get the setup on test, with the option to return if unsatisfactory – because some consequences/constraints are unexpected and to a degree it’s a personal experience. I can very well see this working marvelously in one studio and being a total no-go in another. Now, I’m not getting the financial angle – to me a thin client is an added cost – it would not replace any workstations or make them any less redundant. The idea of a thin client is that the heavy lifting is done elsewhere – workstation, server, on the cloud,... If you mean using a thin client (as in: a very low specced computer) instead of a workstation – that’s something else altogether. Now, a thin client today might more powerful than a supercomputer of the past – so there might be cases where it would work. But if you want to get bang for buck, I’d look elsewhere – as a thin client is not made to customize and beef up and ultimately to put decent specs in. I’d look at barebones rather. From: Angus Davidson Sent: Wednesday, June 10, 2015 8:29 AM To: softimage@listproc.autodesk.com Subject: Virtual Apps Hi Folks Has anyone had any experience using 3d apps like Softimage, Maya, Unity (for our games side) via a thin client. Most of the marketing stuff pretty much makes it look like Christmas in July, but is very thin on actual metrics. ie Latency, numbers of concurrent apps etc. We are in a position where our currency is dropping against the dollar/euro a lot faster then we are allowed to raise fees. So in 2 years when our currently negotiated lease for our 100+ machines runs out, we are looking at the very real possibility of not been able to afford machines that are 3D capable themselves. Oddly one of the few relatively untapped budgets is Major Capex (minimum $50 000 - $250 000), which wont cover a computer but would cover something like a few NVIDIA VCA's Kind regards Angus This communication is intended for the addressee only. It is confidential.
Re: OT:houdini stamp UV
Right click on the node and select, Save texture UV to image. Never used it myself, let us know how it goes! Cris On 10 June 2015 at 09:21, Gerbrand Nel nagv...@gmail.com wrote: Hey guys. My dumb ass is asking dumb questions again. I cant find something similar to stamp uv in houdini Is it there, and how and where? Thanks G
OT:houdini stamp UV
Hey guys. My dumb ass is asking dumb questions again. I cant find something similar to stamp uv in houdini Is it there, and how and where? Thanks G
Virtual Apps
Hi Folks Has anyone had any experience using 3d apps like Softimage, Maya, Unity (for our games side) via a thin client. Most of the marketing stuff pretty much makes it look like Christmas in July, but is very thin on actual metrics. ie Latency, numbers of concurrent apps etc. We are in a position where our currency is dropping against the dollar/euro a lot faster then we are allowed to raise fees. So in 2 years when our currently negotiated lease for our 100+ machines runs out, we are looking at the very real possibility of not been able to afford machines that are 3D capable themselves. Oddly one of the few relatively untapped budgets is Major Capex (minimum $50 000 - $250 000), which wont cover a computer but would cover something like a few NVIDIA VCA's Kind regards Angus table width=100% border=0 cellspacing=0 cellpadding=0 style=width:100%; tr td align=left style=text-align:justify;font face=arial,sans-serif size=1 color=#99span style=font-size:11px;This communication is intended for the addressee only. It is confidential. If you have received this communication in error, please notify us immediately and destroy the original message. You may not copy or disseminate this communication without the permission of the University. Only authorised signatories are competent to enter into agreements on behalf of the University and recipients are thus advised that the content of this message may not be legally binding on the University and may contain the personal views and opinions of the author, which are not necessarily the views and opinions of The University of the Witwatersrand, Johannesburg. All agreements between the University and outsiders are subject to South African Law unless the University agrees in writing to the contrary. /span/font/td /tr /table
Re: ICE team went to Maya (2011)
You don't have to work with houdini the way you describe. Even though there is a procedural nature to the application, you don't have to make every single asset you do a super tool. It's up to you, the software doesn't force you into any workflow. If you know you will be sharing an asset with other members of a team then yes this approach makes sense. In regards to the tutorial you mention, from Keith Johnson. He in fact approaches the task on a very softimage way. From all the tutorials I've seen, this one actually doesn't get bogged down on proceduralism. https://vimeo.com/122274907 https://vimeo.com/125116427 Best, Cris On 10 June 2015 at 16:59, Jason S jasonsta...@gmail.com wrote: Indeed that's why I corrected myself, having recalled SideFX (also) having their own publicized SI team which referred to workflow and not ICE which indeed leaves not much to be desired functionality-wise for Houdini except maybe also the 'workflow' or general ease-of-use of ICE... (despite also being not always easy for everyone).. But speaking of workflow, there was a recent Video from someone adopting Houdini, which I think shows exactly both the main strength and the main drawback of Houdini at the same time. Making a neat and quite detailed retro neon sign, with lightbulbs, neon fixtures and all that, at the end he could dynamically change the amount of fixtures, or dynamically change basically anything. But the time (and complication) involved setting everything up.. in my opinion is worth it if you need a bunch of similar but different things, being faster to redo different variations once the setup is there. And although the fact that you could also do an almost entirely (or partly) procedural dynamic sing (or whatever) setup in XSI is besides the point, (being perhaps less, but also very non-destructive, perhaps like what AfterEffects is to Nuke at least for the stack) the point is you don't -have- to keep everything dynamic if you don't need to, and get things done in a jiffy with much less head scratching or chin rubbing , and you can then say, ok next? That while having quite a bit of things that really don't take any more or less time making (or leaving) them dynamic or not, and involves somewhat minimal history stack management (which otherwise you barely know it's there until you need it or want to clear it) if and when needed to remain fully procedural, perhaps with compounds with exposed params or custom parameters driving operator properties and such (for quick editing in unified property sets -if- necessary). usually being a pretty tiny percentage of things, like we end-up freezing most of everything that doesn't need to remain live once their done. Making it not only a good compromise between fully procedural and non-procedural, but also a best both worlds in many respects. But I guess Houdini doesn't have something that drives many Houdini users away... which doesn't matter if you're at a place who doesn't care about 'dead' labels while waiting for better things, or for things to become better. That being said, even when only looking at what can work best, Houdini of course obviously (also) has it's own strengths uses and place. For Fabric, although I think adopting a visual development environment is a step in the right direction, I personally would hope they would move still quite a bit further away from being a mostly for TD's thing, and would also feel more confident if it wasn't subject to what host DCC's allows to be accessed being probably widely varying from a DDC to another concerning it's scope or reach. say if it could be fully unrestrained in it's own standalone environment, cause ICE apart geo and rig processing, can (very interactively) get set pretty much anything in a given scene, weight maps, uv's, CAV's, particles/strands, materials, any animatable parameter can be ICE driven or drive other ICE things, ... and above-all, all in a very (yet relatively) non-'overtechnical' setting. I guess time will tell. cheers, On 06/09/15 23:09, Raffaele Fragapane wrote: Should probably add, modelling too, and you can pretty much swap animation and modelling in my mail above and all of it remains pretty much true :) On Wed, Jun 10, 2015 at 1:07 PM, Raffaele Fragapane raffsxsil...@googlemail.com wrote: They never said anything involving ICE AFAIK. They did mention that they are trying to make Houdini more pleasant to use, and want animation (the user facing, artist oriented part of it) to receive more attention, and that they have a Softimage developer on board contributing to that. Both those statements are true. They are doing good work in those regards, and part of that work is done by an ex Softie. To my knowledge that's the extent of it. I've yet to see or hear ICE mentioned at all, let alone ex developers of it, by SideFX people. On Sat, Jun 6, 2015 at 5:40 PM, Andy Goehler
Re: ICE team went to Maya (2011)
Indeed that's why I corrected myself, having recalled SideFX (also) having their own publicized SI team which referred to workflow and not ICE which indeed leaves not much to be desired functionality-wise for Houdini except maybe also the 'workflow' or general ease-of-use of ICE... (despite also being not always easy for everyone).. But speaking of workflow, there was a recent Video from someone adopting Houdini, which I think shows exactly both the main strength and the main drawback of Houdini at the same time. Making a neat and quite detailed retro neon sign, with lightbulbs, neon fixtures and all that, at the end he could dynamically change the amount of fixtures, or dynamically change basically anything. But the time (and complication) involved setting everything up.. in my opinion is worth it if you need a bunch of similar but different things, being faster to redo different variations once the setup is there. And although the fact that you could also do an almost entirely (or partly) procedural dynamic sing (or whatever) setup in XSI is besides the point, (being perhaps less, but also very non-destructive, perhaps like what AfterEffects is to Nuke at least for the stack) the point is you don't -have- to keep everything dynamic if you don't need to, and get things done in a jiffy with much less head scratching or chin rubbing , and you can then say, ok next? That while having quite a bit of things that really don't take any more or less time making (or leaving) them dynamic or not, and involves somewhat minimal history stack management (which otherwise you barely know it's there until you need it or want to clear it) if and when needed to remain fully procedural, perhaps with compounds with exposed params or custom parameters driving operator properties and such (for quick editing in unified property sets -if- necessary). usually being a pretty tiny percentage of things, like we end-up freezing most of everything that doesn't need to remain live once their done. Making it not only a good compromise between fully procedural and non-procedural, but also a best both worlds in many respects. But I guess Houdini doesn't have something that drives many Houdini users away... which doesn't matter if you're at a place who doesn't care about 'dead' labels while waiting for better things, or for things to become better. That being said, even when only looking at what can work best, Houdini of course obviously (also) has it's own strengths uses and place. For Fabric, although I think adopting a visual development environment is a step in the right direction, I personally would hope they would move still quite a bit further away from being a "mostly for TD's" thing, and would also feel more confident if it wasn't subject to what host DCC's allows to be accessed being probably widely varying from a DDC to another concerning it's scope or reach. say if it could be fully unrestrained in it's own standalone environment, cause ICE apart geo and rig processing, can (very interactively) get set pretty much anything in a given scene, weight maps, uv's, CAV's, particles/strands, materials, any animatable parameter can be ICE driven or drive other ICE things, ... and above-all, all in a very (yet relatively) non-'overtechnical' setting. I guess time will tell. cheers, On 06/09/15 23:09, Raffaele Fragapane wrote: Should probably add, modelling too, and you can pretty much swap animation and modelling in my mail above and all of it remains pretty much true :) On Wed, Jun 10, 2015 at 1:07 PM, Raffaele Fragapane raffsxsil...@googlemail.com wrote: They never said anything involving ICE AFAIK. They did mention that they are trying to make Houdini more pleasant to use, and want animation (the user facing, artist oriented part of it) to receive more attention, and that they have a Softimage developer on board contributing to that. Both those statements are true. They are doing good work in those regards, and part of that work is done by an ex Softie. To my knowledge that's the extent of it. I've yet to see or hear ICE mentioned at all, let alone ex developers of it, by SideFX people.
Re: ICE team went to Maya (2011)
On 9 June 2015 at 23:07, Raffaele Fragapane raffsxsil...@googlemail.com wrote: They never said anything involving ICE AFAIK. Neither did Jason. They did mention that they are trying to make Houdini more pleasant to use, and want animation (the user facing, artist oriented part of it) to receive more attention, and that they have a Softimage developer on board contributing to that. Both those statements are true. They are doing good work in those regards, and part of that work is done by an ex Softie. Maybe factually true, but meant to mislead IMHO, because they don't have a background UI/animation/modelling workflows, they come from XSI's mental ray integration and viewport HD. That makes these developers equal to just about any experienced developers starting into something new. IMHO it's quite different when ADSK or Fabric says, ho, we're working on something procedural and we have these guys who worked on something similar before in XSI.
Re: ICE team went to Maya (2011)
Well they already have pretty much what Fabric and Bifrost wants to be. They can say whatever they want! ; D I mean its not like Sidefx is going to wait idle when fabric/bifrost grows up. Houdini engine anyone? On 9 June 2015 at 23:07, Raffaele Fragapane raffsxsil...@googlemail.com wrote: They never said anything involving ICE AFAIK. Neither did Jason. They did mention that they are trying to make Houdini more pleasant to use, and want animation (the user facing, artist oriented part of it) to receive more attention, and that they have a Softimage developer on board contributing to that. Both those statements are true. They are doing good work in those regards, and part of that work is done by an ex Softie. Maybe factually true, but meant to mislead IMHO, because they don't have a background UI/animation/modelling workflows, they come from XSI's mental ray integration and viewport HD. That makes these developers equal to just about any experienced developers starting into something new. IMHO it's quite different when ADSK or Fabric says, ho, we're working on something procedural and we have these guys who worked on something similar before in XSI.
Re: GDC presentation: Guilty Gear Xrd and Softimage
@Eugene, would it be too much trouble to show a clip of video of what you´re mentioning ? (a combo of unfold then walking on mesh quad regular quad or projection seems to do the trick, didn't even have to click any corners it was auto!)- Thanks! :D On Tue, Jun 9, 2015 at 6:43 PM, Eugene Flormata eug...@flormata.com wrote: thanks guys! a combo of unfold then walking on mesh quad regular quad or projection seems to do the trick, didn't even have to click any corners it was auto! -- Portfolio 2013 http://be.net/3dcinetv Cinema TV production Video Reel https://vimeo.com/3dcinetv/reel2012
Re: Virtual Apps
In terms of workstation costs, HPDell may be frustrating to get into when on a limited budget. The entry fee is already pretty high and any basic feature will add up quickly to a really, really hefty pricetag. It´s not all their fault, Intel has pricetags that are pretty much unrivaled, not just for Xeons. It´s really hard to build reasonably priced Xeon based workstations nowadays. My DualXeon MacPro2008 was incredibly cheap for what it could do... The good news is, there is now less need for a Dual Socket, Dual Xeon 12Core, workstation machine with a quadro this or that card really when learning 3dComp. It´s nice to have ultra fast machines of course but a decent Quadcore or Sixcore or 8core Single Socket Machine with a good Geforce 9xx and at least 32GB RAM (more is better) will be a nice machine to work with and render with. Add a system ssd and its solid for less than 1650 EUR. Maybe even including a reasonably good display, like a mid-price Dell. Not the fastest to have but a good educational experience for learning to use a renderer properly, too. How about finding an alternative vendor for a lease? Have a reliable (local) shop build and maintain the machines, find a way to make the leasing less than three years, get a batch of refurbished machines to start with and have those run only 2 years, with a return contract or trade in type of deal? You could still try to get some money/funds for a render server with loads of bandwdth and diskspace? That could last 3 years? Cheers, tim Am 10.06.2015 um 17:44 schrieb Angus Davidson: Dear Peter Thank you for the incredibly comprehensive response. The crazy kindergarten accountancy at the university means that the lab computers need to be paid for by the schools from their operating budgets (which are not keeping up with inflation). However things like VCA are expensive enough to be considered Major Capex and that amazingly enough they have funds for. So its mostly about reading the situation at the University and trying to plan around it. Kind regards Angus *From:* pete...@skynet.be [pete...@skynet.be] *Sent:* 10 June 2015 02:41 PM *To:* softimage@listproc.autodesk.com *Subject:* Re: Virtual Apps if you mean using a thin client on the desk, to connect to a remote workstation (in the server room) – then yes – have used this at a former studio. overall it worked quite well. on the thin client you would launch an app, on which you chose the workstation to login to and then a full screen window opens on which you see the workstations’ desktop – and you work you session. It’s very intuitive – apart from a few keyboard combos (ctrl-alt-del is on the thin client, so there’s a different combination to send that to the workstation) You could use the thin client at any desk to log in to any equipped workstation – handy at times – chaotic when your team members end up all over the place. The overhead on the workstation is pretty much zero. The added card handles the compression/communication – so you can push the workstation exactly as before. there was hardware compression/decompression of all signals – so it meant adding a dedicated card in the workstation - all data (kb, mouse, usb as well as monitors) goes through network. afaik the screen refresh is done on the thin client – which reduces the amount of data to be sent (no screens full of pixels) but also makes sure that despite long cable length, image quality is high . (compared to all KVM extenders I ever saw) To the very demanding artist there is a barely noticeable lag and some degradation – you can kind of make out the compression – but you do have to look for it. We decided on using the thin clients only for 3D artists, not for compositors. It would work for compositing most of the time, but when checking final images/shots, occasional little flicks or spots from the compression are disturbing. If you are the person who has 3 oversized monitors on his desk, and expects to have film quality visuals while modeling – this might not be for you. image quality can suffer from network load – as compression adapts some – and at a few peak moments network was so taxed (not because of the thin clients) that connections between clients and stations were lost massively. That’s unfortunate and real disruptive – but once the load was balanced again you would just login and the workstation was right where you left off – preferable to crashes and shutdowns. But it’s something to be aware of - if you have a problematic network, thin clients will add to the frustration. An added benefit was that there was much less heat generated and electricity used in the office rooms – in small cramped, badly ventilated and badly equipped offices that can be a tangible benefit. I have memories of humming workstations under desks, burning desklights and running ventilators
Crowd effects (ICE) - 2011 / 2015?
Hi, I just happen to cross with these videos from AceMastermind1 https://www.youtube.com/channel/UCf5Ojl3TZBnHeGMlT0vLHsgand I was wondering if there´s any new setup tutorial on Crowd behavior for SI 2015? Reading help manual is an aid, but it´s better if it´s on video and showing setups like this one: https://www.youtube.com/watch?v=TE43eZz7bgE Any new crowd effect tutos? I´ve seen all videos with new features from SI 2014 and the official AD videos with basic setups... Is there any deep -official-ICE-crowd training material to look for? Thanks. David. -- Portfolio 2013 http://be.net/3dcinetv Cinema TV production Video Reel https://vimeo.com/3dcinetv/reel2012
Re: GDC presentation: Guilty Gear Xrd and Softimage
yeah what Fabian said! On Wed, Jun 10, 2015 at 2:31 PM, Fabian Schnuer Gohde list@gohde.no wrote: @Pierre http://download.autodesk.com/global/docs/softimage2014/en_us/userguide/index.html?url=files/tex_editonobj_CreatingSubprojections.htm,topicNumber=d30e591106 last paragraph explains how to apply it to component selections via the texture editor and the auto Eugene is refering to is simple right clicking when it prompts you to click corners, it will more often than not figure it out by itself and give you a nice rectangle, i recommend making sure maintain aspect ratio ist checked. For a video check Softimage Contour Stretch on youtube, there should be some stuff there. You will find the operator in the projection cluster and there you can change modes and options. I just used this the other day unwrap a bunch of pipes with bevels and all in the corners and it saved the day. -Fabian On 10 June 2015 at 22:37, Pierre Schiller activemotionpictu...@gmail.com wrote: @Eugene, would it be too much trouble to show a clip of video of what you´re mentioning ? (a combo of unfold then walking on mesh quad regular quad or projection seems to do the trick, didn't even have to click any corners it was auto!)- Thanks! :D
Re: GDC presentation: Guilty Gear Xrd and Softimage
@Pierre http://download.autodesk.com/global/docs/softimage2014/en_us/userguide/index.html?url=files/tex_editonobj_CreatingSubprojections.htm,topicNumber=d30e591106 last paragraph explains how to apply it to component selections via the texture editor and the auto Eugene is refering to is simple right clicking when it prompts you to click corners, it will more often than not figure it out by itself and give you a nice rectangle, i recommend making sure maintain aspect ratio ist checked. For a video check Softimage Contour Stretch on youtube, there should be some stuff there. You will find the operator in the projection cluster and there you can change modes and options. I just used this the other day unwrap a bunch of pipes with bevels and all in the corners and it saved the day. -Fabian On 10 June 2015 at 22:37, Pierre Schiller activemotionpictu...@gmail.com wrote: @Eugene, would it be too much trouble to show a clip of video of what you´re mentioning ? (a combo of unfold then walking on mesh quad regular quad or projection seems to do the trick, didn't even have to click any corners it was auto!)- Thanks! :D On Tue, Jun 9, 2015 at 6:43 PM, Eugene Flormata eug...@flormata.com wrote: thanks guys! a combo of unfold then walking on mesh quad regular quad or projection seems to do the trick, didn't even have to click any corners it was auto! -- Portfolio 2013 http://be.net/3dcinetv Cinema TV production Video Reel https://vimeo.com/3dcinetv/reel2012
Re: GDC presentation: Guilty Gear Xrd and Softimage
and I thought MODO had innovation on that subject. LOL. SI had it first! :D On Wed, Jun 10, 2015 at 4:31 PM, Fabian Schnuer Gohde list@gohde.no wrote: @Pierre http://download.autodesk.com/global/docs/softimage2014/en_us/userguide/index.html?url=files/tex_editonobj_CreatingSubprojections.htm,topicNumber=d30e591106 last paragraph explains how to apply it to component selections via the texture editor and the auto Eugene is refering to is simple right clicking when it prompts you to click corners, it will more often than not figure it out by itself and give you a nice rectangle, i recommend making sure maintain aspect ratio ist checked. For a video check Softimage Contour Stretch on youtube, there should be some stuff there. You will find the operator in the projection cluster and there you can change modes and options. I just used this the other day unwrap a bunch of pipes with bevels and all in the corners and it saved the day. -Fabian On 10 June 2015 at 22:37, Pierre Schiller activemotionpictu...@gmail.com wrote: @Eugene, would it be too much trouble to show a clip of video of what you´re mentioning ? (a combo of unfold then walking on mesh quad regular quad or projection seems to do the trick, didn't even have to click any corners it was auto!)- Thanks! :D On Tue, Jun 9, 2015 at 6:43 PM, Eugene Flormata eug...@flormata.com wrote: thanks guys! a combo of unfold then walking on mesh quad regular quad or projection seems to do the trick, didn't even have to click any corners it was auto! -- Portfolio 2013 http://be.net/3dcinetv Cinema TV production Video Reel https://vimeo.com/3dcinetv/reel2012 -- Portfolio 2013 http://be.net/3dcinetv Cinema TV production Video Reel https://vimeo.com/3dcinetv/reel2012
Re: ICE team went to Maya (2011)
Of the lot probably only Fabric can lay a heritage claim of sorts I believe, not that it really matters though. All in all this is taking an odd turn, and I think part of it stems from a general misunderstanding of development by people who don't do it for a living. I like how you are the arbiter of who's a real ICE developer, from your perspective as a user in Australia and not involved with the development in Montreal. Anyway, the point was only about that animation workflow bit being a bit of a stretch imho. The Fabric or Bifrost/ICE heritage.. well that's a bit more nuanced discussion than claims laid over developers, because AFAIK neither group is trying to simply re-create ICE. Bifrost is a group of developers from various background, some knowing nothing of either ICE or Maya, but getting Oscars for their fluid work (http://tinyurl.com/o9vatz5), old people, new people, plus a group people from Softimage including ICE core developers, caching/SDK/modelling/etc, QA,doc,etc. But they're not really trying to re-create ICE or make a descendant of it, what they're saying is that the design of the graph workflows (still to come) will be informed by it. There is not a single day at work where someone doesn't say we did this that way in ICE because ... But there are also a lot of ideas coming from other places, background, and production experiences. On 10 June 2015 at 19:26, Raffaele Fragapane raffsxsil...@googlemail.com wrote: I wasn't implying Jason did :) I was just saying I don't know of SESI ever mentioning ICE developers coming on board. As for the rest, I don't think it was meant to be misleading, but sure, people will make some tenuous logical connections at times and might have misled themselves making them. By the same standards loosened only the tiniest bit then AD's claims about ICE dev are also somewhat weak, since the people who did the actual heavy lifting, design and architectural work are, AFAIK, not the ones that ended up staying 'til today (speaking of ICE, not Naiad). This is where it starts getting uncomfortable as it nears name dropping, and I'm not comfortable with that. There might have been 20 people tagged in relation to ICE, but at the heart of it there are what, 4 or 5 names that really made a difference? And of those one has left CG and one has left Software vendors (or had last time I checked, which was a while ago).
Re: Crowd effects (ICE) - 2011 / 2015?
I support this, does anyone knows a good in depth tutorial about it? On Wed, Jun 10, 2015 at 12:13 PM, Pierre Schiller activemotionpictu...@gmail.com wrote: Hi, I just happen to cross with these videos from AceMastermind1 https://www.youtube.com/channel/UCf5Ojl3TZBnHeGMlT0vLHsgand I was wondering if there´s any new setup tutorial on Crowd behavior for SI 2015? Reading help manual is an aid, but it´s better if it´s on video and showing setups like this one: https://www.youtube.com/watch?v=TE43eZz7bgE Any new crowd effect tutos? I´ve seen all videos with new features from SI 2014 and the official AD videos with basic setups... Is there any deep -official-ICE-crowd training material to look for? Thanks. David. -- Portfolio 2013 http://be.net/3dcinetv Cinema TV production Video Reel https://vimeo.com/3dcinetv/reel2012
Re: ICE team went to Maya (2011)
I wasn't implying Jason did :) I was just saying I don't know of SESI ever mentioning ICE developers coming on board. As for the rest, I don't think it was meant to be misleading, but sure, people will make some tenuous logical connections at times and might have misled themselves making them. By the same standards loosened only the tiniest bit then AD's claims about ICE dev are also somewhat weak, since the people who did the actual heavy lifting, design and architectural work are, AFAIK, not the ones that ended up staying 'til today (speaking of ICE, not Naiad). This is where it starts getting uncomfortable as it nears name dropping, and I'm not comfortable with that. There might have been 20 people tagged in relation to ICE, but at the heart of it there are what, 4 or 5 names that really made a difference? And of those one has left CG and one has left Software vendors (or had last time I checked, which was a while ago). Of the lot probably only Fabric can lay a heritage claim of sorts I believe, not that it really matters though. All in all this is taking an odd turn, and I think part of it stems from a general misunderstanding of development by people who don't do it for a living. On Thu, Jun 11, 2015 at 3:57 AM, Luc-Eric Rousseau luceri...@gmail.com wrote: On 9 June 2015 at 23:07, Raffaele Fragapane raffsxsil...@googlemail.com wrote: They never said anything involving ICE AFAIK. Neither did Jason. They did mention that they are trying to make Houdini more pleasant to use, and want animation (the user facing, artist oriented part of it) to receive more attention, and that they have a Softimage developer on board contributing to that. Both those statements are true. They are doing good work in those regards, and part of that work is done by an ex Softie. Maybe factually true, but meant to mislead IMHO, because they don't have a background UI/animation/modelling workflows, they come from XSI's mental ray integration and viewport HD. That makes these developers equal to just about any experienced developers starting into something new. IMHO it's quite different when ADSK or Fabric says, ho, we're working on something procedural and we have these guys who worked on something similar before in XSI. -- Our users will know fear and cower before our software! Ship it! Ship it and let them flee like the dogs they are!
RE: Virtual Apps
Dear Peter Thank you for the incredibly comprehensive response. The crazy kindergarten accountancy at the university means that the lab computers need to be paid for by the schools from their operating budgets (which are not keeping up with inflation). However things like VCA are expensive enough to be considered Major Capex and that amazingly enough they have funds for. So its mostly about reading the situation at the University and trying to plan around it. Kind regards Angus From: pete...@skynet.be [pete...@skynet.be] Sent: 10 June 2015 02:41 PM To: softimage@listproc.autodesk.com Subject: Re: Virtual Apps if you mean using a thin client on the desk, to connect to a remote workstation (in the server room) – then yes – have used this at a former studio. overall it worked quite well. on the thin client you would launch an app, on which you chose the workstation to login to and then a full screen window opens on which you see the workstations’ desktop – and you work you session. It’s very intuitive – apart from a few keyboard combos (ctrl-alt-del is on the thin client, so there’s a different combination to send that to the workstation) You could use the thin client at any desk to log in to any equipped workstation – handy at times – chaotic when your team members end up all over the place. The overhead on the workstation is pretty much zero. The added card handles the compression/communication – so you can push the workstation exactly as before. there was hardware compression/decompression of all signals – so it meant adding a dedicated card in the workstation - all data (kb, mouse, usb as well as monitors) goes through network. afaik the screen refresh is done on the thin client – which reduces the amount of data to be sent (no screens full of pixels) but also makes sure that despite long cable length, image quality is high . (compared to all KVM extenders I ever saw) To the very demanding artist there is a barely noticeable lag and some degradation – you can kind of make out the compression – but you do have to look for it. We decided on using the thin clients only for 3D artists, not for compositors. It would work for compositing most of the time, but when checking final images/shots, occasional little flicks or spots from the compression are disturbing. If you are the person who has 3 oversized monitors on his desk, and expects to have film quality visuals while modeling – this might not be for you. image quality can suffer from network load – as compression adapts some – and at a few peak moments network was so taxed (not because of the thin clients) that connections between clients and stations were lost massively. That’s unfortunate and real disruptive – but once the load was balanced again you would just login and the workstation was right where you left off – preferable to crashes and shutdowns. But it’s something to be aware of - if you have a problematic network, thin clients will add to the frustration. An added benefit was that there was much less heat generated and electricity used in the office rooms – in small cramped, badly ventilated and badly equipped offices that can be a tangible benefit. I have memories of humming workstations under desks, burning desklights and running ventilators everywhere (including on an opened workstation case which is a very bad idea) creating an unpleasant and unhealthy microclimate. The switch to thin clients was heavenly. As were LED desklights. Hope it helps some. It’s a big step – that you need to consider carefully with your supplier (ours was HP) – and ideally in a riskfree way, where you get the setup on test, with the option to return if unsatisfactory – because some consequences/constraints are unexpected and to a degree it’s a personal experience. I can very well see this working marvelously in one studio and being a total no-go in another. Now, I’m not getting the financial angle – to me a thin client is an added cost – it would not replace any workstations or make them any less redundant. The idea of a thin client is that the heavy lifting is done elsewhere – workstation, server, on the cloud,... If you mean using a thin client (as in: a very low specced computer) instead of a workstation – that’s something else altogether. Now, a thin client today might more powerful than a supercomputer of the past – so there might be cases where it would work. But if you want to get bang for buck, I’d look elsewhere – as a thin client is not made to customize and beef up and ultimately to put decent specs in. I’d look at barebones rather. From: Angus Davidsonmailto:angus.david...@wits.ac.za Sent: Wednesday, June 10, 2015 8:29 AM To: softimage@listproc.autodesk.commailto:softimage@listproc.autodesk.com Subject: Virtual Apps Hi Folks Has anyone had any experience using 3d apps like Softimage, Maya, Unity (for our games side) via a thin client. Most of the marketing
Re: ICE team went to Maya (2011)
UmI'm assuming you're referring to Halfdan and Guillame? I cannot say for sure, but I'm tempted to say Halfdan has some workflow and animation experience from his days at Pison prior to joining Softimage. Matt Date: Wed, 10 Jun 2015 13:57:42 -0400 From: Luc-Eric Rousseau luceri...@gmail.com Subject: Re: ICE team went to Maya (2011) To: softimage@listproc.autodesk.com On 9 June 2015 at 23:07, Raffaele Fragapane raffsxsil...@googlemail.com wrote: They never said anything involving ICE AFAIK. Neither did Jason. They did mention that they are trying to make Houdini more pleasant to use, and want animation (the user facing, artist oriented part of it) to receive more attention, and that they have a Softimage developer on board contributing to that. Both those statements are true. They are doing good work in those regards, and part of that work is done by an ex Softie. Maybe factually true, but meant to mislead IMHO, because they don't have a background UI/animation/modelling workflows, they come from XSI's mental ray integration and viewport HD. That makes these developers equal to just about any experienced developers starting into something new. IMHO it's quite different when ADSK or Fabric says, ho, we're working on something procedural and we have these guys who worked on something similar before in XSI.
Re: ICE team went to Maya (2011)
You might be taking this a bit more personally and at heart than I thought you would to be honest. I don't consider myself the arbiter of anything, but I guess I'll get out of this since it seems people get touchy easily. On Thu, Jun 11, 2015 at 11:36 AM, Luc-Eric Rousseau luceri...@gmail.com wrote: Of the lot probably only Fabric can lay a heritage claim of sorts I believe, not that it really matters though. All in all this is taking an odd turn, and I think part of it stems from a general misunderstanding of development by people who don't do it for a living. I like how you are the arbiter of who's a real ICE developer, from your perspective as a user in Australia and not involved with the development in Montreal. Anyway, the point was only about that animation workflow bit being a bit of a stretch imho. The Fabric or Bifrost/ICE heritage.. well that's a bit more nuanced discussion than claims laid over developers, because AFAIK neither group is trying to simply re-create ICE. Bifrost is a group of developers from various background, some knowing nothing of either ICE or Maya, but getting Oscars for their fluid work (http://tinyurl.com/o9vatz5), old people, new people, plus a group people from Softimage including ICE core developers, caching/SDK/modelling/etc, QA,doc,etc. But they're not really trying to re-create ICE or make a descendant of it, what they're saying is that the design of the graph workflows (still to come) will be informed by it. There is not a single day at work where someone doesn't say we did this that way in ICE because ... But there are also a lot of ideas coming from other places, background, and production experiences. On 10 June 2015 at 19:26, Raffaele Fragapane raffsxsil...@googlemail.com wrote: I wasn't implying Jason did :) I was just saying I don't know of SESI ever mentioning ICE developers coming on board. As for the rest, I don't think it was meant to be misleading, but sure, people will make some tenuous logical connections at times and might have misled themselves making them. By the same standards loosened only the tiniest bit then AD's claims about ICE dev are also somewhat weak, since the people who did the actual heavy lifting, design and architectural work are, AFAIK, not the ones that ended up staying 'til today (speaking of ICE, not Naiad). This is where it starts getting uncomfortable as it nears name dropping, and I'm not comfortable with that. There might have been 20 people tagged in relation to ICE, but at the heart of it there are what, 4 or 5 names that really made a difference? And of those one has left CG and one has left Software vendors (or had last time I checked, which was a while ago). -- Our users will know fear and cower before our software! Ship it! Ship it and let them flee like the dogs they are!
