Jump to content

DanGrover

Members
  • Posts

    189
  • Joined

Personal Information

  • Country
    United Kingdom

DanGrover's Achievements

Newbie

Newbie (1/14)

10

Reputation

  1. Also check your global switches - it's easy to turn lights off!
  2. We get around this problem at work by placing a batch file in the startup folder, rather than the actual shortcut itself. The batch file deletes all the mapped drives, re-add's them, opens an explorer.exe in each of those locations, then starts up server.exe. The reason we open an explorer is that we were finding that sometimes the drives would remap but have a red X. All you needed to do was double click them to go straight into the drive (And thus remove the red x) but no applications could access those drives until you did so. Opening explorer.exe in those drives just automates that manual process of double clicking the drive. I love Windows.
  3. This licensing cost is actually one of the big things stopping us moving to VR3. Almost all of our rendering is done in the cloud, and we have, on occasion, had 400 render nodes cooking away at our frames. That'd cost us just shy of £50k.
  4. Yup, this is my favourite thing - this works especially well in animations where the camera is moving past or over or panning around the glass - it almost ends up looking like playing cards flipping over when the reflections change from reflecting something bright to reflecting somewhere in the shade.
  5. Snow, Rain etc are pretty much the "Hello World!" of particles - I'm sure, if the presets don't fill your boots, that you won't struggle to find a tutorial or two with a bit of googling; I don't know any off the top of my head, however.
  6. That's true Justin, and it certainly goes without saying that there's a given point at which in-house hardware will be cheaper -if, of course, you can guarantee you'll use it X amount before the hardware becomes obsolete (and of course that's only part of it - the hardware on, for ex, Amazon is updated on a semi-regular basis with the costs not necessarily going up either). We have little in the way of dedicated local render capacity - but we have about 14 workstations that get added to the farm at night, most of which are dual Xeon monsters. Of course, if we can render using these we always will. The times we use Amazon, though, are if we need a particularly quick turnaround (if you have a 400 frame sequence at an hour a frame, our little-farm-that-could would take approximately 22 solid hours, assuming no one was using their workstations - Amazon can do it in just over an hour) and it's in these situations that we find that flexibility being the most useful. If we had this sort of throughput constantly then it's certainly true that it's cheaper to buy the hardware. That said, as a fairly small, agile company, there's something to be said for our entire network infrastructure being able to fit into a small, 20U rack!
  7. You could try doing a separate render with a material over-ride. This could just be a VRay Light material (adjust for exposure) with a falloff map in the diffuse. You can set the falloff map to use "Distance" in the z-axid with a near and far value and two colours (which, conveniently, default to black and white). I believe this will work with an orthographic view/camera.
  8. PFlow? I mean, it's a super simple process to just dump some particle's down from way up high with a slow downwards speed (I wouldn't use gravity - whilst technically more realistic, it'll keep increasing in velocity if you do this which you probably don't want for snow) and a very soft, slightly turbulant wind knocking it to the side. You can set the particle's to just "stop" when they hit the deflector (floor) and use a "stop gradually" to have them gently rotate to the floor, where they'll stay. Of course, for the sake of efficiency you'll probably want there to be a decent looking snow material on the floor already. But if you dump enough particles on the floor, you can cover up anything!
  9. Hi All, Whilst doing the 2014 CGArchitect Survey (which you should all do if you haven't already because it's great) there's a question about where the industry is heading, and it's a text-input option - ie, not multiple choice. This makes it fairly difficult to quantify, and whilst I'm sure Jeff will diligently find a way via an arduous slog through our inane ramblings, I thought it might be good for us to have a place to discuss and possibly show-and-tell what we done wrote. Where do you see the industry heading in the next 5 years? I think staying on full CPU (as opposed to GPU) rendering but with it all hooked up nicely to the cloud. As broadband speeds and provisions increase, the "troubles" with cloud rendering with become less and less of an issue. When competition really begins in the space, I think people will stop looking at GPU rendering as being a cheaper (insomuch as time = money) alternative to CPU but rather they'll concentrate on quality, which at the moment tends to be higher on CPU thanks to fewer constraints. I also think that in-house viz at architectural practices will start to eat the low and even mid tier viz companies' breakfasts - in house teams might be fairly expensive but they're a) cheaper than high end companies and b) far quicker for prototyping than the lower end, cheaper (typically internationally outsourced) companies. The quality and speed of desktop machines and rendering software (for ex VRay and it's entirely decent sun and sky model are available for more and more software platforms inc architectural ones) that I think the low end part of the industry that's typically used for competition/bidding projects or design-and-development projects will find little to offer practices. Furthermore, the gradual increase in the use of cloud rendering will reduce the need for expensive, up-front local render farms, further increasing the potency of small, in house teams. I think that what will ultimately remain are the more high end studio's whose dedication to the process (that is, the ones able to properly study the whole pipeline for potential improvements, develop in house tools and art styles, and who's typically marketing-focus allows for more visually interesting films and images) will be the only one's who will hold out longest in the battle to be replaced by small internal teams. What recent advancements do you see as being most significant to the advancement of the architectural visualization industry. Cloud rendering, without a doubt. I think it's a real game changer - it's one of the few "hurdles" of a CG film process that we can really throw money at. If we want a render quicker, we can spend more. There aren't many parts of the process where this is the case, as even hiring more people doesn't necessarily result in a faster outcome (too many cooks etc - which isn't an accusation anyone ever used against a render farm!) It allows us to work for longer before hitting render which improves quality. It allows us to respond more quickly to feedback from clients, and it allows us to be more transparent with our clients when it comes to charges for re-renders that result from requested changes that fall outside the original scope (we can say with certainty, honesty and - if necessary - proof that a certain shot cost us, say, £200 to render, so that can be added onto the bill for any changes required). This isn't limited to the ArchViz industry, but it gives us a larger degree of flexibility compared with before, and its uptake will go up and its price down in the future, I think. ---------- That's what I wrote, anyway. What do you guys think? I think this is a pretty broad discussion and not the first time we've done it, but I thought it was a timely reboot. If Jeff has a problem with this discussion - insomuch as we're spilling some of the beans from the survey - then of course close it up, but hopefully it's fine since it's basically the least useful part of the survey as it's just our (admittedly somewhat educated) predictions as opposed to quantifiable data.
  10. I'm a bit late to the party here, but I just wanted to add (and I think this is C G's point, so I apologise if I'm doubling up here) that I think the point re: privileges etc is that you're (most likely - visas permitting etc) free to move to Bangladesh etc where you, too, can live for much cheaper than you can in Norway. Then you too could do as you're suggesting these people do and still charge a lot for your images (after all, your talent won't have reduced) and then you could live a life of relative luxury in the third world whilst maintaining high levels of cost for images. But you haven't - now this could be because of family ties, because you like the life in Trondheim or a combination of both. But there are lots of people that aren't from London currently living in London and working in the Arch Viz industry, and I daresay they've chosen to come here due to a combination of the large pool of work and jobs available, plus the fact it's a nice city with modern infrastructure and whatnot. The point being that for us, in the west, we need high prices because we couldn't afford to live here otherwise. We could move to Bangladesh and then we wouldn't need to, but we don't because we like the privileges and comfort that comes from living in the west, and which also cost us a great deal of money. That's why these privileges are relevant - because they're something that most aren't willing to jettison.
  11. I have a photo of me lying on a sofa with my girlfriend's lovely dog lying on my lap. I figure that for every potential opportunity my ugly mug loses me, the dog gains me one, so it's a wash.
  12. Honestly, the issues I have with rendering our visualisations and other animations is never the volume of geometry, because we already have tricks to keep the footprints down (Proxies, instances, plugins etc). My problem is always things like AA settings, glossy reflection samples, frosted glass - basically shading. It's shading that "holds me back" insomuch as that's really where I have to make compromises now, not geometry. Maybe that's all because those things do, in part, help mask the fact that the geometry is polygonal, but it's still a problem. I think for games this could (Assuming they sort out the lighting and whatnot) have massive implications. For what we do, I'm less sure.
  13. Hi, Holding down left click over the dotted square symbol button two buttons to the left of the move tool in the toolbar allows you to choose various ways of selecting, including lasso. And I believe the quad-divide tool you're looking for can be done a few ways. On a per-polygon level, try hitting the Tesselate button within edit poly. Over the entire mesh, if you add a 'Subdivide' modifier, this should have the same effect. Or the tesselate modifier. Or the Quadify Mesh modifier. The options are endless and confusingly overlapping! Edit: Damn my having the window open for 20 minutes!
  14. There are a million and one things this could be!! Firstly, when you render, try clicking the little circle in the render window that's half black and half white. This shows up the alpha - white means there's geometry there (even if you can't see it), and black means there's nothing there. If it's all white, then it's probably a lighting problem. If it's all black, then it's some kind of problem with the geometry. This might not fix the problem, but if you can let us know we might be able to help troubleshoot it further from there!
×
×
  • Create New...