Autodesk acquires Solid Angle

[Bullit]

Btw what happened to Lagoa cloud renderer, was also bought by Autodesk?

[Unaided]

Anyone knows a opinion from inside Whiskytree? Cos I have a feeling as that company. First, kick to your pipeline base software and now other kick to its main render engine.

Personally I need now migrate to other render engine for Softi. Arnold is great renderer but I don't wanna have any relationship with AD (beyond SI, that I consider NO adesk sw). And in the other hand SItoA devel is dead.

Maybe is time to I should move to GPU render based. Redshift or Octane seems good alternatives. However I will need invest my money in new hard.

Other option that I thinked is RenderMan, but I think only 3delight is RM for SI, is it? No Pixar compliant exists for SI? And 3delight support is currently active for SI?

And a last thing. Would not be implicated AD in Panama's offshore accounts scandal and go bankrupt? At least by this way 3D industry, 3D market and 3D uses could living in peace.

[MauricioPC]

I read an article at 3DArtist (the latest issue) on how companies are using Redshift. Glassworks for example switched from Arnold to Redshift and they LOVE it. They said there very little to almost nothing that Redshift can't do and Arnold can.

[Maximus]

Good thinking i moved away from AD. I'll never give them any cent anymore.
Gladly moving to Redshift and Maxwell.

I have no idea how people can slightly think anything good would come out of this, Marcos and solidangle can say whatever they want, the facts and the past speaks for itself, alot louder than simple marketing bullshit.

Anyway not that i care, im out from AD roof since a while
I just feel bad for people who are under it and forced to eat this crap everytime, arnold users. Also think for a minute, arnold is developed also for softwares not autodesk, do you really think autodesk will let the same development be on a competitor? like c4d? nnnnnnnnnnno ;p


waiting on the announcement on 2 november this year when AD will buy Chaos Group

[Unaided]

Chaos Group was save the star product of AD when this was throught bad times. V-Ray was the salvator for 3DSMax. I have no doubts.

I have two opposite thinkings about AD and Chaos Group.
In the side one, AD knows that the big part of VRay users are Max users. This is a 'life insurance' for AD. Therefore AD no need to struggle for it.

In the side two, AD had attempt many times acquire Chaos Group, but due to disagree in negotiations AD no had had success. I would thinking the cause is Chaos Group no wants sell under nothing money amount.

[nDman]

...moving to Redshift and Maxwell.

That's what I do too, Redshift for fast render and Maxwell for super realism.

However, I'm forced to use Renderman and Arnold in workplace.

[Unaided]

3delight is not good option to replace SItoA, for personal projects that involving animation sequences?

I think that Renderman in the past was born as scanliner and its strong points was fast displacements, but RM had been famous to needed write shaders to obtain maximun quality. I want to say, RM is or was the opposite of Arnold in terms of artist-friendly use.
Recent versions of Renderman have more simplicity to obtain good results in short time? It have raytracing, photons o its point-cloud based GI method is enough? I don't find a fully PBR but the simplicity vs. results of Arnold or Octane are important things actually.
3delight and Pixar Renderman they have important diferences?

[Hirazi Blue]

3delight for Softimage sadly stopped being a true alternative,
as they quite apparently have stopped investing much time in it after the EOL announcement,
not even fixing the obvious bugs in their most recent free version...

[luceric]

just so we're clear, I think gpu rendering s incredible.

the point I poorly made earlier was, try to put yourself in the shoes of a company who wants to get into a cloud rendering service. You need 1) a proven renderer that people want to use. Not some no name or second/third renderer that you need to sell people into using 2) something that distribute well to hundreds of inexpensive clusters 3) a user base that does something which requires a lot of processing power, and hopefully increasingly prohibitive to do locally.

Otoy is *already* a cloud rendering service company, one which isn't widely used. People have the option to use it right now, and by large that's not what studios are sending their big work to. What they are actually doing is using arnold render farms and Amazon clould solutions. Even google is offering that https://www.zyncrender.com/

[FXDude]

For rendering, if what you do in life is render, then mostly renting render time is probably not a good idea.

But can still be good for customers for extra needs that happen on occasion. (but as an only, or as an extra pushed option?)

For example a 1 X 600 frame (10 sec) render job at a conservative 30 min per frame on a given Cloud service can easily cover a Redshift licence, and another job (or 2) like that covers a pretty decent GPU that can render 600 frames a day round the clock all year until it burns-out, and or is eventually replaced by another faster one, and a couple of those can be put into one box. You can make tests all you want, don't have to confom to farm idiosyncrasies/supported features, transfer textures, voxel or mesh cache files(typically heavy), and download full frames with all AOVs + ( perhaps 50-75 meg Deep files?) mounting to easily dozens of gigs for regular exrs which even if technically possible with mega connections I find it just silly for regular usage.



But the probem with cloud services, can be the same as PAAS SAAS or any other AAS (RAAS?)

It's popular among providers (with -all- marketing around it) because time or space is sold in much smaller chunks, which cost less for the customer up front and -can- be less trouble (and more trouble in other ways), but are consistently and considerably more expensive (like renting a car, or your tools as a carpenter?, or going to the restaurant every day) and thus proportionately more profitable not too far down the line, and it involves extra dependencies which is probably precisely the point, or I don't think one can trust that yet further dependance won't (further) be taken advantage of, with plenty of examples of that up to rental only policies.

[emphasis mine]

From a Vendor perspective, if your customer can buy subscription, then subscription definitively makes more sense:

1. Stable cash flow instead of rolling the dice every Q

2. Constant and continuous growth that will please the board, investors and markets, rather than the Q roller coaster driven by your market seasonality or suboptimal sales team effort repartition.

3. You usually can charge more over the relationship. Customer who value the flexibility and cash flow improvement [improvement in rather specific circumstances of temporary necessity] of a subscription model will pay a premium compared to a perpetual license

4. Customers have less negotiation power:
the relationship usually start with a lower cash outlay and is less likely to attract purchasing department attention [smaller regular payments] or intense competition. As your service and billing grow, Customers might come back to negotiate but then the fact that they will have to incur cost and risk (i.e. "switching cost") to unplug your solution, will put you in a much better negotiation position. Not mentioning that they will also have seen and measured the value you deliver before talking money (ie you and they know they need your gears).


5. You might avoid RFP because the initial transaction is below the threshold that triggers it.

There are plenty of other good reasons to favor a subscription model as a vendor. Even if some of these will disappear over time, as buyer get more sophisticated and get a better understanding of this type of negotiation [ie: 'customer awareness']

Pesky customer awarness..

[NNois]

just so we're clear, I think redshift is incredible.

the point I poorly made earlier was, try to put yourself in the shoes of a company who wants to get into a cloud rendering service. You need 1) a proven renderer that people want to use. Not some no name or second/third renderer that you need to sell people into using 2) something that distribute well to hundreds of inexpensive clusters 3) a user base that does something which requires a lot of processing power, and hopefully increasingly prohibitive to do locally.

Otoy is *already* a cloud rendering service company, one which isn't widely used. People have the option to use it right now, and by large that's not what studios are sending their big work to. What they are actually doing is using arnold render farms and Amazon cloud solutions. Even google is offering that https://www.zyncrender.com/

Sure, the best way for Arnold is to promote cloud rendering, this work only on CPU and we are sort of falling under Moore law, yeah... render faster ? pay more !
You can add to that the today trend to dish workstation for smaller clients, maybe in some years you won't find workstations at all (Apple already started that).

I wish to think also that, ok, with a big processing power and "inexpensive clusters" you can render faster but you are also sucking earth ! by the way that's important, what's the greenest calculators CPU or GPU ?

[angus_davidson]

From What I can see GPU will win out in the end. Simply look at the increase in CPU speed vs GPU Cuda cores

[bb3d]

Here is an interesting statement from Panos Zompolas (one of the Redshift developers):

Hello everyone,

Recently there were a few threads in our forums discussing the Autodesk-Solid Angle deal. Given that some of them contain phrases like “please, don’t sell to Autodesk”, we thought we should comment on this.

Like some of you, we first heard rumors of this deal several weeks/months ago. Anyone can speculate on why Marcos decided to sell his company but we’d rather not do that - at least not publicly! smile

The main question is whether this deal will affect Redshift in any way. Some of you probably think that if Arnold eventually gets bundled with Maya, this might mean fewer sales for Redshift. Well, since the day we released our first non-free build, there have been several occasions where our competitors slashed prices or even (temporarily) offered their product for free! None of these actions affected our sales - they kept rising. Also let’s not forget that there are already free products out there - like Blender/Cycles. That is to say that we don’t think this deal will make much of a difference to Redshift’s sales. In fact, since Autodesk’s announcement, sales have gone through the roof! Maybe that’s pure coincidence, maybe not. It’s sometimes hard to extrapolate data in this market.

The topic of Arnold GPU rendering is something that was also mentioned recently. I will repeat here the same answer I gave several months ago: don’t hold your breath! smile Developing a GPU renderer like Redshift is not trivial. Don’t assume that a company developing a (great) CPU renderer will simply be able to port their technology to the GPU. GPUs don’t work that way.

Another topic is the Autodesk-ZYNC deal and how this could affect GPU rendering in general or Redshift in particular. One of the toughest projects a studio could get (from a rendering standpoint) is full-CG episodic content. I.e. having to render many hours of full-CG animation in a short amount of time. You’d think that renting cheap CPU blades would be the obvious choice here. Without breaching any NDAs, I can tell you that Redshift is currently being used by several studios that work on such large-scale full-CG productions. These studios evaluated cloud rendering but eventually chose to render in-house with Redshift. A price of a dollar an hour (or less) sounds great until you realize how many dollar-hours you’ll have to pay for in order to complete your project. What about using the cloud for interactive work? Well, imagine if pencils and brushes were rented by the hour. I wonder how much time people could afford to spend practicing their art.

Having said all that, cloud rendering offers some obvious benefits. Sometimes studios need a temporary boost in rendering power. While we’re not quite ready to discuss this in any useful extent, there are some exciting opportunities ahead for Redshift. We hope to be able to talk about this in the relatively near future. But the main point here is that we don’t think the cloud will revolutionize rendering in quite the way some people expect or hope - at least not in the short term. There are significant technical, logistical and legal obstacles ahead.

Finally, I’d like to address the common “please don’t sell Redshift to Autodesk” comment often seen in our forums. You can rest assured, we’re currently in no talks with anyone about such a thing. Our plan remains the same as it has always been: to keep developing Redshift, expand it to more 3d apps, increase the audience’s awareness and continue expanding its market share.

Thanks to everyone for their continuing support!

-Panos

[MauricioPC]

Loved tis reply. Good stuff.

[juanloredo]

Just sharing this:

Marcos Fajardo on the Arnold renderer and Autodesk’s recent acquisition

http://www.3dvf.com/actualite-16817-int ... derer.html

[BenR]

Just sharing this:

Marcos Fajardo on the Arnold renderer and Autodesk’s recent acquisition

http://www.3dvf.com/actualite-16817-int ... derer.html

Thanks. Even though 3DVF is a French page, "L'interview est en anglais, sous-titrée en français."