How does one read a crystal ball
John Bond
Posts: 369
Congratulations on your acquisition of the SX range of micro-controllers. It has been apparent for some time that Ubicom lacked a vision for the line,
·
A few months ago Elector, the European electronics magazine, suggested that you should try to stick with one micro controller because you are better able to fully exploit the device’s resources if you understood it well. For me, the transition from Stamp to SX isn’t being too difficult and the SX’s raw processing power has its advantages. I am therefore a new SX user (very new). Clearly, if you are entering the market without Stamp experience, you would need a good reason not to go with the market leader (Microchip) but I already know and use the BS2p.
·
I am concerned about my investment in time to learn the SX language. The competitors are offering fancy features such as 10-bit A to D, PWM etc. etc. Yes, you can do all that with the SX but, in my eyes, the product no longer has the edge. There has been no development of the silicone chip and there have been no additional hardware features. In my humble opinion, this is mainly due to Ubicom’s lack of vision for the product but, if I’m going to switch, now is the time to do it.
·
Parallax has traditionally been very strong in strategy. Forget Dell, MartinAir, Wallmart and Harley, the companies that the Ivy League Business Schools traditionally push as experts on strategic vision. You have passion for your product and you understand your users. You have well segmented your market, and you sell a couple of products that are intrinsically the same to different market segments at vastly different prices. You grow your market (literally) by introducing the product at school. Your customers are generally as excited about your products as you are. As the Zulus would say “Eyesh! Parallax has vision”.
·
So, without giving too much away, what is your vision for the SX range?
Kind regards from Kwa Dukuza
John Bond
·
A few months ago Elector, the European electronics magazine, suggested that you should try to stick with one micro controller because you are better able to fully exploit the device’s resources if you understood it well. For me, the transition from Stamp to SX isn’t being too difficult and the SX’s raw processing power has its advantages. I am therefore a new SX user (very new). Clearly, if you are entering the market without Stamp experience, you would need a good reason not to go with the market leader (Microchip) but I already know and use the BS2p.
·
I am concerned about my investment in time to learn the SX language. The competitors are offering fancy features such as 10-bit A to D, PWM etc. etc. Yes, you can do all that with the SX but, in my eyes, the product no longer has the edge. There has been no development of the silicone chip and there have been no additional hardware features. In my humble opinion, this is mainly due to Ubicom’s lack of vision for the product but, if I’m going to switch, now is the time to do it.
·
Parallax has traditionally been very strong in strategy. Forget Dell, MartinAir, Wallmart and Harley, the companies that the Ivy League Business Schools traditionally push as experts on strategic vision. You have passion for your product and you understand your users. You have well segmented your market, and you sell a couple of products that are intrinsically the same to different market segments at vastly different prices. You grow your market (literally) by introducing the product at school. Your customers are generally as excited about your products as you are. As the Zulus would say “Eyesh! Parallax has vision”.
·
So, without giving too much away, what is your vision for the SX range?
Kind regards from Kwa Dukuza
John Bond
Comments
My own experience with learning indicates that the SX-Ubicom provides are far more versitle entry point than the PIC of Microchip and Microchip will bury your learning in tons of documentation. For the person with imagination, Guenther has written a quite comprehensive book that will take you a long way.
For the PIC, I suppose that having dozens if not hundreds of differrent chips offers someone some kind of security against people reverse engineering your software code as you have to know what exactly the hardware is, but having to select and work around all those different chips seems to be a bigh waste of time.
Having said that...
Microchip is an excellent resource for seeing the many diverse applications that have actually been brought to market and I do take a peek to try and understand what can really be done.
Just for starters, the PICmicro DC Motor Control 'Tips 'n Tricks' is a great quick and simple read for anyone trying to survey motor control.
Regarding limitations real and not so real, you can add on an SPI or I2C chip to do most of what competitors package and these are coming out so fast that it is hard to keep up. You may find that having one vendor trying to do it all will put you into second place if a SPI or I2C really rocks with innovation.
If anything, the SX-Ubicom needs to expand the amounts of RAM and ROM on board. And of course, there is always the 16 bit micro-controller to be made. I suspect they know this and have been look for ways to go there in a business-like fashion.
I completely agree with sticking with one microcontroller for learning, but you still should look at the competition for both ideas and innovative applications - after all they are looking at you.
Regarding your investment time in learning SX language - it is nearly the same as Microchip PICs. I suspect once you understand the SX fully, you can always program a PIC [noparse][[/noparse]but might not really want to step down to slower speed or the lack of virtual perepherals]. I would be more concerned if you wanted to master Amtel's AVR RISC with 110 instructions [noparse][[/noparse]how can they call that a 'reduced instruction set?].
Regards from topical Taiwan
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
G. Herzog in Taiwan
So why go with the SX? 1) speed, 2) price and most importantly 3) support, go take a look at thier forums http://forum.microchip.com/ I personally find it a bit disappointing, many of the sub catagories have just a few posts, plus I don't see many (any?) microchip employees participating in the discussions. Sometimes questions cannot be answered by someone outside of the company. Plus I find thier software functional but a bit difficult to work with.
When I decided on a platform to develop my ideas on (8 bit), I looked at pretty much all of the major manufacturers (Atmel, Ubicom, Microchip, Rabbit, Renassas, Zilog, Toshiba, Intel, Motorola, etc) and the SX was the best valued general purpose 8 bit microcontroller. This investigation started 2 years ago culminating in purchasing the SX platform just over a year ago, and there has not been a single day that I have regretted making the decision. You may conclude differently, but if you already have the SX platform, I would stick with it (for one most of the other microcontrollers require a substantial investment in third party software to develop in any other language besides assembly, if you are familiar with stamps, the PBASIC-like format of SX/B is a very convenient way to slowly wade into assembly rather than having to take the complete plunge and risk drowning in confusion).
As a side note, I purchased an atmel 8 pin microcontroller (was going to use it as a multi-purpose extension chip), the thing is still sitting on my desk, and never been programmed, and Im quite proficient in assembly. But the whole process is time consuming and not very intuitive.
Ubicom does not lack vision, just go take a look at thier IP line of microcontrollers (the succesor to the SX line of microcontrollers), that are in a word amazing, effectively imitating a 8 way multi-core processor through zero-wait state task switching. Its an impressive line of microcontrollers. But they have marketed the line to large corporations willing to plunk down large amounts of money on development platforms, plus they do all thier sales through distributors. I tried to get the platform when I had more change in my pocket, and the distributor representive was discourteous, dismissive, and downright rude when he found out I was an independent inventor and could not gaurentee upfront that I would be purchasing 10K+ units of the processor. Thats when I said to heck with it and decided to go with·the SX.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·1+1=10
Post Edited (Paul Baker) : 8/29/2005 4:40:17 PM GMT
I am a little concerned about·Parallax taking over the production of the SX chips. (Actually, the Parallax switch and SXlist going away at the same time bothers me a bit!)· I like the support a lot, and I like the chips a lot. They are great. It's just a nagging feeling that the 'end' is near. Everything that I know about parallax tells me that the chip is going to be around for a long time, but I still have this nagging feeling.
-Dan
·
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Jon Williams
Applications Engineer, Parallax
As per the IP series, while you may have applications which require more processing power than the SX can provide, more than 90% of hobbist's applications can be handled by an SX (just think of how many applications Stamp users come up with and those operate·in thousands of operations per second). If Parallax were to carry the IP line, it would be a low volume product, like the Javalin or Altera FPGA platforms. And Parallax does have a bottom line to consider, the resources to support it, the warehouse space to stock it and other incidental costs of expanding the number of processors they carry makes the prospect of carrying the IP not a very attractive prospect for them. If you are truely serious about the IP line, try to procure it through Ubicom's distributor and lie if nessesary to get them to treat you with enough respect to provide it to you. I can't remember what the development platform costs, but it was either in the $500, $800, or $1000 range, and good luck getting any bare processors in sub-reel quantities, and you'll also be on your own as far as support (Ubicom isn't going to provide you with personal support unless you are a major manufacturer).
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·1+1=10
However, There is still probably a bit of market in small companies that need long term support for bigger processors. It doesn't mean that the programmer is $100. A $500 or $1000 programmer may be fine those applications.·(If support is there). Third parties do make money at it developing IDE's. And,·Parallax has a history with Ubicom that could prove benefical·in·such a·venture.··I could Imagine Ubicom marketing a "Parallax" supported IDE...···· ...or how about a Visual Basic·Stamp...·· ....I could spend all sorts of Parallax's time and money if they'd let me!
If I am forced into bigger processors, I'm going to look at a bunch of the 32 bit chips out there (DSP's and otherwise). The problem is that I need a processor that has a very very very long life cycle. I deal with low volume (<25/yr) stuff. My biggest problem is product obsolescence. Just about every week I have to find a source for an obsolete part, or redesign something older because of stuff changing. It's·a big part of the reason I'm using the SX now. (It's also the reason I'm·pushing the powers that be to·allow me to stock reels of components!)
If parallax supported·bigger guns, I'd always be able to get ammo for them! It would make my life much easier! (Even though, I'd probably still have that nagging feeling, Jon!)
-Dan
·· I totally feel your pain there.· My quantities were low volume too as a designer, however the companies I did design for expected a design they could get parts for down the road.· When they couldn't they came back to me.· One major manuafcturer of silicon put me through the ringer several times before I learned my lesson.· Funny thing is that the last chip we designed around, is still around!· And that was from 1995!
·· Anyway, you can rest assured we're making the best possible choices for ourselves, and our customers.· We don't want to leave anyone hanging, including ourselves.· And keeping the SX line going in it's current mature state doesn't mean we're not working on bigger and better things.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Chris Savage
Parallax Tech Support
csavage@parallax.com
You are asking about our business plan. Our vision for the SX is the following:
Not on·our list·of goals are the technical comparisons·between AVRs, PICs, etc. and their internal·hardware. We always aim to present our products based on their own technical merits. And the SX has no internal hardware, so selecting it from a long list of processors with built-in hardware is not a good marketing approach. Instead, our marketing tools are support, Parallax tools, a free compiler, low-cost proto boards, etc. Many of our customers will waste the many SX MIPS available, and that's just fine if you want to run it at·4 Mhz with the internal oscillator. Heck, it's just like a BASIC Stamp in many ways.
Sincerely,
Ken Gracey
Parallax, Inc.
Ok, I look at the "About Parallax" and see the name Chip Gracey. Who is this Ken Gracey? (big grin and lots of respect here).
Thank you Ken. My experiences with Parallax have always been positive. It's amusing to call the office up there and, because the company is small but very efficient, you can talk directly to the department manager and get the issue resolved right then and there.
I'm finding the SX intriguing (I better, I just bought a bunch of them) and I'm having fun tinkering with several of Parallax's sensors and stuff.
Ken, do you think that you could come up with a nice clean and inexpensive download that will turn an SX into a web server ... with I2C communication capability (I know, I'm dreaming, but someday I'll figure out how to get TWO led's to light on my SX chip AT THE SAME TIME!!!)
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
John J. Couture
San Diego Miramar College
Chip is my brother and runs our R&D; I manage the·daily operations and·product marketing. Chip isn't active on the forums, but today I asked him to read this "crystal ball thread".
No plans on turning the SX into a web server.·The SX's resources memory resources were·quite limited once the TCP/IP stack was running, which led Ubicom into the IP2K/3K designs. However, we've got a new partner in the "microcontrollers to internet" business and they're doing a fantastic job putting together exactly what we need to provide·a friendly BASIC Stamp/SX interface to the web (e-mail send/receive, web pages, etc.). The hardware from this effort should be less than 30 days away. Those in need of our prior Red-I device will be pleased.
Okay, what's the problem with your two LED project?
Yours,
Ken·Gracey
Thank you for the peek into the future. It keeps us guessing and salivating over the possibilities of new products.
The corporate title explains your avuncular "voice of authority" that sometimes comes through on your posts in other forums and belies your High School graduation picture above (grin).
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
John J. Couture
San Diego Miramar College
It's interesting that your users, partners and customers took the time to add to·and amplify on your comments. Was it Peter Drucker, that old management guru of the 60’s and 70s (of the last century) who stated that one should strive to create a sense of ownership of your business among your stakeholders be they suppliers, staff or customers? ·Well… ·seems like you’ve done that!
Kind Regards from that place somewhere in darkest Africa.
John Bond