topic Re: What would you pay for speed in Collaboration with other software

What would you pay for speed

Anonymous — Thu, 21 Dec 2006 23:00:40 GMT

Here is a general question
if you could get a rendering package that is capable of real time rendering
(in other words it will produce renderings equal to the best C4D rendering
in a rate of about 15 frames per second) while all materials, shaders, radiosity, HDRI sttings ETC is done in real time so you could see the changes instantly
and the whole thing will work on you're regular work station

How much would you pay for it?

Re: What would you pay for speed

Anonymous — Thu, 21 Dec 2006 23:19:05 GMT

I think that I would pay about $750 for something like that.

Re: What would you pay for speed

Dwight — Fri, 22 Dec 2006 03:37:37 GMT

Most users think that it should be included.

Re: What would you pay for speed

Anonymous — Fri, 22 Dec 2006 23:32:53 GMT

Dwight wrote:
Most users think that it should be included.

Yes in a perfect world
ArchiCAD will work in full render mode all the time
But this question relates to a stand alone product that potentially can work with ArchiCAD
and so if you could potentially produce you're presentations in a mater of seconds what kind of saving it will mean to you
and hence
what would you pay for speed

Re: What would you pay for speed

TomWaltz — Sat, 23 Dec 2006 00:37:04 GMT

Are you planning on selling us something?

Re: What would you pay for speed

Anonymous — Sat, 23 Dec 2006 03:43:36 GMT

TomWaltz wrote:
Are you planning on selling us something?

Not yet.

In future? ........ possibly!

But still it is interesting to find out
(considering the amount of time it takes to create a good presentation)
What is a product that can actually save time, and just minutes, but hours or days or weeks, will worth for the potential buyer?

Re: What would you pay for speed

TomWaltz — Sat, 23 Dec 2006 13:28:36 GMT

I think it would need to be more than just "speed". It would have to be "productivity and quality."

Trying to make the rendering process faster hbas two aspects: artistic/aesthetic quality and processing power.

You seem so to be saying that your potential product would speed the artistic/aesthetic end of it, which I have to say sounds nice but also a little unlikely. With greater power comes greater need for flexibility and varying tastes, which is extremely difficult to automate without Matrix-style mind jacks.

For professional renderers, they will want a LOT of flexibility. For your normal architect, I think they will want to be able to set it up to render in the style they currenlty use. You will need the flexibility in the program to give them what they want.

You see it a lot with Archicad. For some people with certain needs, it's realy great. For people who want to model like Frank Gehry, it's not the best product.

Some people complain that Autocad is not full-featured enough. The bigger problem is that they are trying to make a product that works for architects, engineers, manufacturers, and industrial designers. As a result, you have a great drafting program that's really just an awesome electronic drafting board, but not great for any one application.

Personally, I have Lightworks and C4D already. Anything new will have to be really amazing for me to purchase a new product, learn its workflow, and go through another learning curve.

Re: What would you pay for speed

Dwight — Sat, 23 Dec 2006 17:00:25 GMT

I apologize for being cynical while you are trying to develop a new product.

I said what i did because i'm cheap, like most Archicad users are. We've spent a lot of money on an overall excellent application that has many second-rate features, many of which are excoriated on this forum. I initially mis-spelled "excoriated" as "excoritated." Besides inventing a new word, this describes the feeling, say, of being told that you are getting LightWorks included and discovering it is the old LightWorks not the new LightWorks because that would be too much work. Excoritated.

For a fair answer, then, to get real-time rendering in Archicad is worth thousands, but only to those few who can apply it and aren't resentful of Archicad's weakness in the rendering area. So, as a marketer, you'd want to assess the market SIZE rather than the unit PRICE. Relying on my experience giving illustration seminars and selling my LightWorks book, you'll be alarmed and disappointed to see how few Archicad users actually make photorealistic imagery of any kind from their models.

I don't know the reason but i suspect:
-- it takes a lot of time to properly set-up a rendering scene.
-- even with your promise of real-time feedback, the technology learning curve/scene refinement curve is immense.
-- users don't have the artistic sense to make satisfying images because they haven't been to art school, film school and professional photographer's school, aspects of which all apply when making architectural imagery.

The buyer would be making professional images (?) This means that your product would need to access superior forms of entourage that Archicad currently doesn't - like animated figures and foliage. The things that Cinema 4D can use, for instance. In part, users export their models to place entourage or engage animation features like wind.

And your product needs to be fast on a real model. A "block of flats," say. Those of us who have been taken in by certain oleaginous (look it up) Hungarian-accented voices demonstrating speed when rendering a mocked-up shack to then discover the grinding mess a real project can be are understandably suspicious of speed claims.

But if you have a product that can seamlessly plug-in to Archicad and give real-time rendering the quality of Cinema 4D with sophisticated material and lighting definitions, why stop at Archicad? There's thousands (hundreds??) of Cinema 4D people out there using outside Renderfarms to do their animations. Now, there's a market!

I'd be happy to brainstorm with you about this.

Re: What would you pay for speed

Anonymous — Sat, 23 Dec 2006 22:36:17 GMT

It is very interesting to read you're replies
Thank you.
especially about the artistic understanding involved in a good presentation.

the reason this topic is presented at the AC Forum is partly the fact that I'm an AC user and the AC community is close to heart
And also Dwaight's book (Light Works In ArchiCAD) which is an amazing book (and funny too).

AC can be viewed as a tool to latterly "Build the structure" in a virtual environment , and therefor get a "second chance" so to speak, to do it better faster and cheaper, prior to building it on site
Finishes are a big part of the cost of a building, and one would do well to plan those carefully, so the ability to truly see what certain finishes will look like can be valuable, plus it can result in an unexpected creativity that can spring from process of simulation itself.
So "to render a model is a kind of study" as well as presentation.
But a truly good rendering takes hours to prepare, and I've been longing for a truly fast rendering solution, that is intuitive.
so I guess what I'm trying to find out is. How important is a tool like that for the community? and what they would be willing to invest in a tool that can change the way they been doing things?
And besides speed and quality image, what are the things that will inspire them, to make that investment, so we can tailor a tool that is optimized for our needs as a community.

Re: What would you pay for speed

Anonymous — Sat, 23 Dec 2006 23:58:37 GMT

Tamir wrote:
How much would you pay for it?

Re: What would you pay for speed

Dwight — Sun, 24 Dec 2006 00:00:47 GMT

See?

Re: What would you pay for speed

Anonymous — Sun, 24 Dec 2006 06:10:38 GMT

I don't think I ought to have to pay *anything* extra, or buy any additional software. I've already paid thousands of dollars for Archicad. Top quality rendering software ought to be built in. Rendering is a critical aspect of the work we do.

Wendy

What you say is true. Most of us bought AC as a complete suite and we expected it to do everything for us (and I believe that this is what Grphisoft where aiming at, when they conceived of AC at the beginning)
But intentions are not enough, and all though the presentation tools are there, they are not satisfying for most of us
I guess it is a little bit like buying a boom box VS buying a full pledged Hi Fi system (in which every component will be a stand alone I.E Pre amp, Power Amp, Equalizer, Etc)
and so AC is the equivalent of a boom box (a good one, but still a boom box) and you want a Hi Fi.
perhaps sperate tools for modeling, drafting and presentation are the way to go (as long as they are seamless).
In any case what I understand from what you are saying; is that you are not happy with the AC rendering tools, but still, you will not invest in a sperate tool for presentation that will alleviate your current pain.
Did I understand you correctly?

Re: What would you pay for speed

Anonymous — Sun, 24 Dec 2006 09:21:16 GMT

Tamir wrote:
What you say is true. Most of us bought AC as a complete suite and we expected it to do everything for us (and I believe that this is what Grphisoft where aiming at, when they conceived of AC at the beginning)
But intentions are not enough, and all though the presentation tools are there, they are not satisfying for most of us

True, but I think we ought to be able to *expect* high end performance from a high end product - especially one that claims to have this functionality.

I'm not saying AC doesn't have this yet, just that in general, I think that high end, fast, accurate rendering *ought* to just be an automatic part of high end products.

Of course, some programs do this better than others because of a variety of factors, so part of what drives purchase decisions is how well various features that one considers essential are implemented, with *any* CAD program.

In any case what I understand from what you are saying; is that you are not happy with the AC rendering tools, but still, you will not invest in a sperate tool for presentation that will alleviate your current pain.
Did I understand you correctly?

Well, yes and no. I'm not good enough at all this yet to feel like I've come anywhere near exhausting AC's capabilities, or to understand when a particular result is caused by the software vs my own skills, so I can't really say that I'm unhappy with AC's rendering tools, especially now with v10. I was responding in a more generic manner to what I thought was a more generic question.

But no, I would *not* invest in a separate tool right now anyways, I don't think. I can't afford to, especially with trying to keep the basic product and add-ons up to date. I also just simply don't have the time to keep learning new products and keeping up on their changing versions. If I should find myself with a client who requires higher end rendering skills than I am capable of providing with the tools I have at hand without spending all of my time trying to learn how to produce them, well, Dwight's available for hire the last time I checked <g>.

For school projects? I'll keep plugging away learning what I can as I go, and produce the best results my skills and the tools I already have are capable of, but I am not going to kill myself for utter perfection when no one else is footing the bill. Unfortunately, it is frowned upon in academia to hire out any part of your projects, or I'd probably farm out things like physical model-building and some CAD modelling, rendering, and documentation tasks anyways, if I could swing the cost.

Wendy

Re: What would you pay for speed

Anonymous — Sun, 24 Dec 2006 18:37:36 GMT

Although, I also think AC should have good rendering engine, I still think much more important part of AC is this one that produce Construction Documents (in every stage - from SD to ASI and RFI). If you will reasonably think how much time you spend on producing documents and rendering, you will probably find out why. Lets not forget that even packages that are designed to produce just images and animation, don't have top notch rendering systems (max, maya, etc). If you need better quality you have to buy external rendering systems (Vray, Renderman, etc...). Lets face it. In small company no one is expecting hiperior quality (no one wants to pay for it) in big companies they have specialized team that is doing just it or they hire other company to make it for them.

Re: What would you pay for speed

Anonymous — Sun, 24 Dec 2006 19:34:59 GMT

I just wanted to remind you that this topic was presented in "Other Products" for a reason.
(Other products that can possibly compliment ArchiCAD)
I've never done a project where renderings were not part of package as well as a complete CD'S (since I've been working for a design build firm)
but that is exactly the point.
Most of us don't invest the time to produce good presentations/renderings simply because we don't have it, and not because we don't want to.
and that is because no matter which rendering packages we will use they are still SLOW. They are slow to render and are complicated to set up, so usually we will render a frame more than once to achieve good results, which means lot's and lot's of time.

And so, let's illustrate this in a metaphor.
We are all going to around in a race truck armed with the latest bicycles, some are a little bit better than others but essentially all bicycles are the same, and now lets say that we can replace our bicycles with motorbikes and shave 90% of our lap time.
Is that worth something? considering it can be done.

Re: What would you pay for speed

Anonymous — Sun, 24 Dec 2006 21:44:54 GMT

Tamir wrote:

And so, let's illustrate this in a metaphor.
We are all going to around in a race truck armed with the latest bicycles, some are a little bit better than others but essentially all bicycles are the same, and now lets say that we can replace our bicycles with motorbikes and shave 90% of our lap time.
Is that worth something? considering it can be done.

Well, sure - theoretically, at least. Provided that a) it doesn't cost a bloody fortune and b) the learning curve isn't so steep that the time for just learning how to do it and keep it updated as necessary substitutes for the current rendering time, at least for too long.

I'm just tired of having to use so many different products already, just to produce whatever I'm already able to get out. Every time AC is updated, I also have to update a whole slew of add-ons, and at some point, I have to update other programs like Photoshop/Adobe CS and formZ as well, just to maintain file format compatibility for the times that I need it. This is not only extremely expensive but very time-consuming, just to *install* all the new versions, never mind learning how to use them and their new features. And that's before I start trying to make sure the installations on *two* computers are the same so I can move the drawing back and forth as required without trouble.

Besides, I'm trying to do *interior* design, not graphic design or graphic arts, so how much perfection in rendering do I really need? Yes, admittedly, part of what drove me into AC in the first place was the promise of being able to render things accurately enough that I can visualize what different materials and furniture selections will actually look like, so it's a tough call. Really being able to see what things will look like, and how different materials will make a space feel *does* require a fairly high degree of photorealism.

But I'm already more of a slave to my tools than I want to be. Having to add yet *another* program in order to be able to do what I want to do, particularly when it ought to already be built in to the base product, is a very frustrating thing to have to contemplate. I just spend more time learning and upgrading software (and system maintenance) than I do actually working already, I think, or at least it often seems like it.

Sometimes I wish I'd just spent more time developing my hand rendering skills, and am half tempted now to just forget about it on the computer and just produce hidden line drawings and render them manually.

Perhaps once I have some paying clients, I'll feel differently, but right now, it's all outgo financially, and school deadlines can't be pushed back just to accommodate learning endless new software as I go.

Wendy

Re: What would you pay for speed

Anonymous — Sun, 24 Dec 2006 22:34:01 GMT

So it seems as if when we can let our imagination fly (in terms of our dream tool)
we are willing to consider the possibility of an investment both in terms of money and time
But since only two replies in direct answer to my question
1.) about $750
2.) if it is that good Thousands of $
where given
I would like to learn a bit more about what will the ultimate rendering engine should be like?
Can you let your imagination fly?

Re: What would you pay for speed

Anonymous — Mon, 25 Dec 2006 20:07:55 GMT

Tamir wrote:
I would like to learn a bit more about what will the ultimate rendering engine should be like?
Can you let your imagination fly?

1. It should be as intuitive and easy to use as a paintbrush, for starters

2. It should include accurate photometric data for light fixtures, and radiosity or a better technology that would render finishes and lighting just like it would really look with those fixtures and with the given sun angle.

3. It should allow for multiple finishes on each side of an object.

4. It should be possible to change the parameters of the applied texture on the fly, without having to stop and go to a whole different menu structure.

Wendy

Re: What would you pay for speed

justdhruv — Tue, 26 Dec 2006 07:10:42 GMT

hi tamir, thats a very interesting question you posed. I hope i don't offend any body here, but nobody has understood the spirit of the question. The question being posed, i feel, is the holy grail of rendering. I was introduced to 3D modelling and rendering in the year 1997. I was working on strata, and they had just released radiosity in release 2.1. I had thought to myself, when would computers be fast enough to do real time radiosity. It been almost a decade and we still aren't anywhere near, even realtime raytracing.

I am not aware of any technology available in the market or being developed which can offer that solution at any price point any time soon. It is also like asking "what would you pay for a flying car?". And i think saying "$25000" would be a very irreverent answer.

Coming back to you question. I think such a solution would be priceless. when it is finally available, production houses would gladly pay upwards of $25000 for it. But for me as an architect, i guess at this point of time i could pay around $5000 (for which i will have to take a finance) for a solution which gives me "Maxwell Render" (sice it supposedly does not have to much setup and is easy to use) type of rendering in real time.

Re: What would you pay for speed

Anonymous — Sun, 31 Dec 2006 23:12:08 GMT

Tamir

Seems to me you are talking about an Artlantis on steroids.

As a long time Artlantis user, I would love high end rendering in real time, taking 1/15 of a second on what now takes me over of 1 hour machine time.

I am perfectly happy with having a rendering dedicated software outside ArchiCAD.

AC I use for conceptualization, data management and technical drawings. Full fledged VB.

Artlantis I use for the pretty pictures, that help me (enourmosly) to sell my ideas.

Unfortunately, knowing how computers work, I believe the holy grail you are promising us is still some time in the future.

Our models are not simple semi-3d game models, but very complex and polygon heavy ones, with easily over one or two million polygons.

So, although I know for a fact we will have that kind of computing power in the future, my guess would be 5 to 10 years (more on the 10 years, but I am waiting to see what the cell fuss is all about).

Now, as I am allowed to dream a bit on new years eve, if you would have this dream software for me tomorrow, with ease of use as in artlantis, real time high end rendering on a regular pc, bugless, integrated with Archicad, I would probably be able to convince myself to depart with some $3000.

Happy 2007