Sorry, bit of Rant but I have to ask...

For Flowcode users to discuss projects, flowcharts, and any other issues related to Flowcode 6.

Moderator: Benj

Post Reply
User avatar
Jay Dee
Posts: 398
Joined: Tue Aug 18, 2009 6:42 pm
Has thanked: 121 times
Been thanked: 154 times
Contact:

Sorry, bit of Rant but I have to ask...

Post by Jay Dee »

Hi Guy,
I was finding it hard to be convinced during the trial period but I none the less I see a few new features to justify the fairly small cost of upgrading to FC6. FC5 has served me very well.

I am certainly not out to criticise an excellent product, just voice a level of concern for great software...becoming bloatware with pretty graphics, when it does not actually aid real world hardware design. Given the huge number of man hours that was required in the development of the 3D modelling and basic kinimatics, is this a sign for the future of Flowcode?

As an industrial user, I have yet to be convinced of the reasoning behind the new 3D modelling side...I cant honestly see what this offers in electronics development other than a pretty picture.
Yet some core components such as the SPI component remain in Legacy form.

I guess your education market is larger than the industrial but as a software tool for non-coding engineers to develop embedded programs and systems, FC its fantastic and I sincerely hope for the product to keep growing in depth rather than in breadth.

Again, I'm not knocking some of the great work you have done but we now have 10 different type of switch! with looks like the MicroSoft Windows road to bloatware.
Kind regards,
John. - I'm now going to post my tech question...and find out if I've been black balled! :) :)

User avatar
JonnyW
Posts: 1230
Joined: Fri Oct 29, 2010 9:13 am
Location: Matrix Multimedia Ltd
Has thanked: 63 times
Been thanked: 290 times
Contact:

Re: Sorry, bit of Rant but I have to ask...

Post by JonnyW »

Hi Jay.

There have been a number of queries around 3D. Perhaps I should put down a train of thought as to (one of) the reasons behind this.

Flowcode v5 used bitmaps to represent components. These are fast for the system to render but are time consuming for someone to hand-draw, especially when animation is concerned (such as a switch).

Flowcode v6 had one major bullet point on the spec - it must have the ability for our users to create their own components.

Whereas bitmaps are fine when there is one dedicated developer drawing them and developing components, and they can be reasonably consistent then, it is unrealistic for our user base to take the time to, say, draw an animated switch, and even more so to think that it will look in keeping with the rest of the product. In truth, Flowcode v6 a year down the line would be a miss-mash of inconsistent and cryptic-looking components.

An alternative is vector graphics. A 2D-vector library would certainly allow our users to draw things easier. Problem is that 2D-vector libraries are, to put it mildly, vast. Just considering a line, you have thickness, dashes, ends, elbows, curve type and so on. 2D vectors are also not quick to draw, so to get anything simulating real time would be a big job.

But 3D vector graphics are actually much simpler to implement, as they tend to be much lighter in weight. if you can build it out of triangles, then it is good to go in 3D. There is also the added benefit that most 3D can be hardware accelerated, making it pretty quick during simulation, and are easy to animate.

Problem is that people see '3D' and think 'complicated'. In reality I believe the reverse is true - I can knock together a new type of switch in Flowcode in under a minute. I could not draw the first frame of the push or lever animation in a bitmap in that time.

You are right, when it comes to parts such as LEDs and switches 3D does not aid hardware design. However, when it comes to a robot arm connected by servos, 3D can be quite a useful tool during simulation. It also opens the door to integrate with the many CAD packages that are currently being upgraded or released with 3D environments.

Right now there is not much in a 3D image of an LCD for our embedded users. However, if we released v6 with the ability to create components in 2D, then decided that for v7, or v8, there was a good reason to go to 3D, the underlying principles behind interaction with the system would be so different that we would be faced with the decision to make everyones v6 components obsolete, or accept we can not make that change. If we decided that you could lay out and produce a multi-layer schematic of a circuit board or wanted to know how big a container would be given a set of existing components, we would have to tel you that sorry, Flowcode is not for you, where as now this sort of thing is a possibility in future versions.

As far as multiple switches go, I fully agree that the components toolbar is getting cluttered. I am looking at putting an option in to hide components from the lists so that if you search 'switches' you might get a long list, but the drop-down menus remain quite light. We have been pushed to create 'real-world' looking components as there are large repositories of 3D models available from manufacturers, and this helps open discussions with them. Personally I would use the dashboard components for debugging and diagnostic if you do not care about the appearance of a component - these are designed for clarity.

I hope this clears up some issues. As an aside, in the next release I have provided the ability to turn off the system panel and solely use the dashboard, with the ability for it to be more 'v5-like'.

Cheers,

Jonny

User avatar
Jay Dee
Posts: 398
Joined: Tue Aug 18, 2009 6:42 pm
Has thanked: 121 times
Been thanked: 154 times
Contact:

Re: Sorry, bit of Rant but I have to ask...

Post by Jay Dee »

Jonny,
Thank you for taking the time to post such a detailed response for those of us asking about 3D.
I totally understand the need for taking a technical direction for the ease of future development and longevity.

I guess it is also a case of 'horses and courses'; Those that want a graphical visualisation to aid developing a product within the minds eye, could really benefit from 3D.

Within my work customers, we do plenty in 3D, from Aero CFD to mechanical design...but when it comes to designing electronics systems, I'm happy with the simulation outputting pretty basic LED statuses and data streams within the new consoles. A graphic representation will always be a very 'course' representation of the final system so does not really justify the time spent producing a crude approximation. I guess in simple terms, people like me are after a code accurate simulation, what it looks like is of fairly low importance.

I possibly need to review the concepts behind components...since me building a different looking graphic for my specific switch just seems ...well... I may be missing something.

For us old skool types, the option to hide/minimise the multiple graphical variation of essentially the same electronic component may be one direction... For my typical devleopment, a switch or ADC only need to be of a generic type...we'll do that side of things in the real world mechanical and hardware design.

My rant aside, lets not forget the great work that has been done in progressing an great tool.
Regards, John.

Unfortunately I've been down the route of entire design projects going 3D and CAD jockies showing me fantastic models, all animated and dont get me going on the 3D wiring harness deisgn software...ha ha.. many many thousands of software that they were sold and screwed the entire wiring installation of a high end automotive project. 3D has its place but so do the old methods.

Post Reply