It looks like a very nice board.  Pity that those of us who run Linux can't play.

It seems so bizarre to me that a company whose main product is semiconductor devices (not software) deliberately reduces its customer base by playing favorites on operating systems.  It's just sheer blinkeredness and incompetence to not write development software in a portable manner so that people can use their chips regardless of which operating system they happen to use.

Vandana, if you're in contact with Cypress, you might like to pass this comment along to them.

Morgaine.

It wasn't done out of spite for other operating systems or anything so diabolical : )  We want to go cross platform, We know its important, but we have man-decades (maybe even man-centuries) of effort in our existing tools, built on a windows platform.  Is it perfect?  Of course not.  Its going to take a lot of work to redo the tools to make them cross platform like they should be.

I guess I just want you to know, we do hear you, and we are working to address it.  Or at least, I hear you : )

You don't need to necessarily rewrite all the tools to be cross-platform (though than would be really nice). Make at least sure they play nicely with Wine. That way we can _run_ them under Linux, even though they are not integrated as I we would wish.

(With the latest version of Crossover, the installer for 2.2SP1 now at least starts, but will crash when showing the license. But at least it seems to get better with each version...)

its built heavily on the .NET framework, so sadly, it probably would take a rewrite to make it work well with wine.  I haven't checked to see how well it works with mono, might be worth a shot.

Chris Keeser schrieb:

 

its built heavily on the .NET framework, so sadly, it probably would take a rewrite to make it work well with wine.  I haven't checked to see how well it works with mono, might be worth a shot.

I can install .NET under Crossover, so that should not be the issue (in fact, the installer checks that in its first stage, and is happy with what it finds...).

(Actually, I just did some further experiments and now the installer seems to be running. I will report what happens...)

yeah, definitely let us know how it goes!

The installer hangs when "caching" the installation files (I don't know what its trying to do there, everything is local...). I will look into it this this evening, amybe the Crossover logfile is helpful to see what is going wrong.

John Beetem wrote:

 

From what I can tell, there really isn't a GUI standard so if you want portability you use X Windows (X11) and hope that the particular distro does a good job of implementing it or provide a reliable adaptation layer for whatever it's using for graphics.

As usual, it's not quite so simple. There are two widely adopted toolkits, GTK & Qt, and the interesting part is that if you write with either of those as the target rather than any native (X11 or ms windows) then there are implementations of both toolkits that run on top of X11, Windows, OSX and possibly more. So from a UI perspective you can develop it once. You do obviously need to be aware of all the odd stuff like case insensitive filenames on windows and not code your app to make life difficult for those sorts of things.

  And I never could figure out how to print directly from a GNU/Linux program, so I just generate a PDF and require the user to print it using "lpr" or an PDF reader.  This is OK for work-in-progress software like XXICC, but you'd like something more seamless for a final product.

Using one of the toolkits helps here as they have already abstracted a lot of the detail for you into a reasonable print dialog that you can use to send your print job without needing to know the detail of how the backend printing system works on that OS.

Plus, if you need to access custom hardware such as a device programmer, you'll get to explore the netherworld of device drivers.

These days, when most devices are USB anyway, it's probably a good idea to build the programmer to use one of the standard CDC classes. If that's possible to do, you likely don't even need to worry about a driver as all OS's will have them already.

Release the documentation for PSoC routing, and the Open Source Software world will do the work for you.  Intel doesn't have to write their own compilers -- others happily do it for them.  ARM doesn't either.  Neither does MIPS or PowerPC.  Xilinx and Altera have chosen to write all their own software, and IMO that's prevented them from being the next Intel or ARM.

One good example is the Atmel AVR stuff that's supported by GCC.

Some companies work with the community and that can be hugely beneficial, other try to work outside it or even actively against it. The results are predictable, people who work with the community get carried along with it and often can even find someone else will keep the code up to date with changes, the ones that work against it find their code frequently broken or get boxed into only one distro.

Opening the documentation on how to use the device is usually the best policy, it generates goodwill, and that might make all the difference.

DAB wrote:

 

Hi Chris,

 

No need to appologize for being windows only.

When you build tools you look at your market potential While there may be 2-3 million loyal Linux/Unix users there are  300-500300-500 million Windows users

The number of users of an OS is almost irrelevant if your target audience is using something else. Historically Apple did well in the DTP area for those reasons. Market potential is much more than the simple number you state.

Ultimately if the PSoC is an interesting enough device then people will use whatever tools are available, if not they'll take the path of least resistance and pick an equivalent that doesn't require an additional OS and probably PC to go with it just to run a single tool.

DAB wrote:

 

When you build tools you look at your market potential While there may be 2-3 million loyal Linux/Unix users there are  300-500300-500 million Windows users

That's only part of the picture though, and says nothing about market relevance.  Borrowing from an earlier post,

Among the many millions of Linux desktop users, the proportion with technical backgrounds or interests is enormously higher than for Windows or Mac.  The Linux/Unix world grew largely through word of mouth within communities devoted to science, technology and higher education, and this has given it a very different demographic to operating systems with a user base in office and media consumption and/or general use.

The "2-3 million loyal Linux/Unix users" that you mention may well contain as many technically-oriented people as your "300-500 million Windows users", and it could be a lot more.  This makes the topic of "market potential" very different to a simple comparison of overall operating system user populations.

Statistically, the average Windows user is extremely unlikely to know what a microcontroller is, let alone an FPGA.  In contrast, the average Linux user is likely enough to recognize at least one of these two terms that it's worth placing a bet on it.  And at the other end of the scale, experts in technology are disproportionately well represented in Linux/Unix circles.

It almost suggests that a more strategic approach in the technical industries would be to support Linux users first, since that would be more efficient targeting of the desired audience.  The desired audience among Windows users is comparatively sparse.

Hi Morgaine,

Thanks for your post.  By the time I finished reading it I was laughing so hard I had tears in my eyes and was gasping for breath.

I have seen a lot of descriptions of Linux users, but yours is by far the most hilarious I have ever read.

I have not laughed so much since my injury over ten years ago.

I look forward to your next post with eager anticipation of further entertainment.

Thanks for sharing,

DAB

I'm glad to oblige, but in the absence of any information or reasoning in your post, it's just an unsubstantiated refutal.

The difference in the demographics is well known and not a matter of dispute.  Unix evolved from deep research and academic roots, and Linux was until fairly recently "by programmers, for programmers".  In contrast, Microsoft has always targeted Windows as an operating system for everybody.  The demographics reflect this.

If you'd had deep exposure to the Unix, Linux, research and educational communities over a long period of time, you might have noted the difference yourself.  Having been part of all of these since the late 70's when I began running early version of Unix on PDP-11 off tapes received from Bell Labs, I've observed the very high proportion of technical people in the Unix communities first-hand.

This has been the case for over 3 decades, and it is only in the last decade that Linux has also become "cool" and started to be used by less technically oriented people.  It all contrasts very vividly with the general population of Windows users, which I have also seen widely through working as a contractor in industry and among friends, where "technical competence" on average means knowing how to use Windows office apps, and more recently a browser.

The O/S demographics are as different as chalk and cheese.  While of course there are lots of technical people who use Windows, they form a tiny proportion of Windows users, whereas the corresponding proportion in Unix/Linux is very high.

This should surprise nobody.  Windows is used by the general population of personal computer users, and therefore reflects the interest and occupation demographic of the general population.  Technical competence is rare in the general population.

I'm glad to oblige, but in the absence of any information or reasoning in your post, it's just an unsubstantiated refutal.

The difference in the demographics is well known and not a matter of dispute.  Unix evolved from deep research and academic roots, and Linux was until fairly recently "by programmers, for programmers".  In contrast, Microsoft has always targeted Windows as an operating system for everybody.  The demographics reflect this.

If you'd had deep exposure to the Unix, Linux, research and educational communities over a long period of time, you might have noted the difference yourself.  Having been part of all of these since the late 70's when I began running early version of Unix on PDP-11 off tapes received from Bell Labs, I've observed the very high proportion of technical people in the Unix communities first-hand.

This has been the case for over 3 decades, and it is only in the last decade that Linux has also become "cool" and started to be used by less technically oriented people.  It all contrasts very vividly with the general population of Windows users, which I have also seen widely through working as a contractor in industry and among friends, where "technical competence" on average means knowing how to use Windows office apps, and more recently a browser.

The O/S demographics are as different as chalk and cheese.  While of course there are lots of technical people who use Windows, they form a tiny proportion of Windows users, whereas the corresponding proportion in Unix/Linux is very high.

This should surprise nobody.  Windows is used by the general population of personal computer users, and therefore reflects the interest and occupation demographic of the general population.  Technical competence is rare in the general population.

The internet is written in ink.

How do you feel about this incredibly embarrassing temper tantrum you threw now?

Please please please fly off the rails again, I haven't been this entertained in the comments section of a product demo ever.