What specs will entice design engineers to investigate the prop ?
Cluso99
Posts: 18,069
Here is some suggestions to open the discussion...
I will use "we" and "our" for Parallax.
One thing I have thought of is like other micros do.· We could have a table of features, or a comparison chart, that lists·each variant of the family of Propeller chips.· We could make some really great tables. We could then make the point that because the peripherals are not tied to specific pins, all our chip variants are pin compatible, all cost the same, and are ACTUALLY the same chip because the peripherals are soft and extremely intelligent. In other words, make the point that the prop has a soft configured peripherals. But we make the point at the end, not the start.
We have to make the point that the peripheral blocks are super intelligent and therefore simple to use. Perhaps we could say they are microcoded peripherals that run in their own dedicated onboard 32bit risc processor. We have lots of precoded configurable peripheral blocks (in the OBEX).
We have to make the point that Interrupts are NO LONGER required. So your code is MUCH simpler as all the intelligence of the peripherals is off-loaded to other dedicated onboard 32bit risc processors. There are 8 identical processors onboard (which includes the main processor).
We all know C is required. I don't think we need to say how C specifically works (LMM I mean). In the end it is going to be by performance comparisons and the prop is not a slouch.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Links to other interesting threads:
· Home of the MultiBladeProps: TriBlade,·RamBlade,·SixBlade, website
· Single Board Computer:·3 Propeller ICs·and a·TriBladeProp board (ZiCog Z80 Emulator)
· Prop Tools under Development or Completed (Index)
· Emulators: CPUs Z80 etc; Micros Altair etc;· Terminals·VT100 etc; (Index) ZiCog (Z80) , MoCog (6809)·
· Prop OS: SphinxOS·, PropDos , PropCmd··· Search the Propeller forums·(uses advanced Google search)
My cruising website is: ·www.bluemagic.biz·· MultiBlade Props: www.cluso.bluemagic.biz
I will use "we" and "our" for Parallax.
One thing I have thought of is like other micros do.· We could have a table of features, or a comparison chart, that lists·each variant of the family of Propeller chips.· We could make some really great tables. We could then make the point that because the peripherals are not tied to specific pins, all our chip variants are pin compatible, all cost the same, and are ACTUALLY the same chip because the peripherals are soft and extremely intelligent. In other words, make the point that the prop has a soft configured peripherals. But we make the point at the end, not the start.
We have to make the point that the peripheral blocks are super intelligent and therefore simple to use. Perhaps we could say they are microcoded peripherals that run in their own dedicated onboard 32bit risc processor. We have lots of precoded configurable peripheral blocks (in the OBEX).
We have to make the point that Interrupts are NO LONGER required. So your code is MUCH simpler as all the intelligence of the peripherals is off-loaded to other dedicated onboard 32bit risc processors. There are 8 identical processors onboard (which includes the main processor).
We all know C is required. I don't think we need to say how C specifically works (LMM I mean). In the end it is going to be by performance comparisons and the prop is not a slouch.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Links to other interesting threads:
· Home of the MultiBladeProps: TriBlade,·RamBlade,·SixBlade, website
· Single Board Computer:·3 Propeller ICs·and a·TriBladeProp board (ZiCog Z80 Emulator)
· Prop Tools under Development or Completed (Index)
· Emulators: CPUs Z80 etc; Micros Altair etc;· Terminals·VT100 etc; (Index) ZiCog (Z80) , MoCog (6809)·
· Prop OS: SphinxOS·, PropDos , PropCmd··· Search the Propeller forums·(uses advanced Google search)
My cruising website is: ·www.bluemagic.biz·· MultiBlade Props: www.cluso.bluemagic.biz
Comments
"Now, you see, this counter mode here counts up whenever the user-defined two input logic equation is satisfied."
"M'kay. Now where'd you say that UART peripheral was?"
-Phil
Now your analogy for the UART... This UART is microcoded (it's an object in the Obex) and it can have variable buffers. It can expand str, hex, dec, bin comands. And it could do a lot more. The engineer does not need to know how a UART works as it is already available as a block, but you can extend its' functionality if you want by adding microcode.
The next is the SD / SPI driver. We have this here micocoded object that not only implements the SPI interface, but it also does the low level read and write to an SD card. Once again, forget what the SPI interface is - you don't need to know, just add the microcoded object from the Obex.
This is where I am coming from.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Links to other interesting threads:
· Home of the MultiBladeProps: TriBlade,·RamBlade,·SixBlade, website
· Single Board Computer:·3 Propeller ICs·and a·TriBladeProp board (ZiCog Z80 Emulator)
· Prop Tools under Development or Completed (Index)
· Emulators: CPUs Z80 etc; Micros Altair etc;· Terminals·VT100 etc; (Index) ZiCog (Z80) , MoCog (6809)·
· Prop OS: SphinxOS·, PropDos , PropCmd··· Search the Propeller forums·(uses advanced Google search)
My cruising website is: ·www.bluemagic.biz·· MultiBlade Props: www.cluso.bluemagic.biz
As I have said before we should drop the language of COGs and HUB etc. Use terms the uninitiated would immediately recognize and feel comfortable with.
Talk about "peripheral blocks" and flexibility not COG's.
HUB RAM is just "RAM"
Such a collection of peripheral drivers, FullDuplexSerial, SD card, I2C, SPI etc define a variant of the Prop "System On a Chip". Same chip, just comes out of the box with that selection of drivers ready to go.
In that respect I think the COG's PASM code should be referred to as "firmware" rather than "microcode".
"Microcode" has a connotation of being very deep within the system and high complex/specialized, something mere mortals do not want to touch.
"Firmware" on the other hand is something people are used to. Just a binary blob that you have to pass into some hardware to make it jump. You don't have to think about it.
Wish I could find my post suggesting the Prop as "System On a Chip" again.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
I think you are crossing threads here. Cluso is talking in this thread specifically about wider adoption by design engineers - who will perfectly well understand the concepts of cogs, hubs, counters, Hub RAM vs Cog RAM, PASM, LMM etc - even if they don't use precisely those terms when communicating with their commercial managers, financial backers or business partners - good engineers are usually quite adept at "tailoring" (I don't want to say "dumbing down" - but you know what I mean) their technical terminology when dealing with "outsiders".
However, you are talking more about adoption at the retail level - in that area I agree that what the Propeller mostly needs is not just a new terminology, but an entirely new approach to development - one that really does allow users to develop applications without requiring a deep knowledge of cogs and hubs etc etc.
To do that it would be nice if you actually could treat things like the OBEX software objects as "blobs" - but you know you can't. Currently, to get these "blobs" to work correctly within a larger application nearly always requires a fairly intricate knowledge of the internal "plumbing" of the Propeller - i.e. those very cogs and hubs that you don't want to mention. We probably don't realize how bad it is because we already have this knowledge. But think how much time on these forums is spent answering questions about how to pass information between cogs, or how to start multiple instances of this or that object (or why you simply can't do so)? For that matter, many times have your own applications failed to run because you have run out of cogs, or miscounted the number of cogs required, or gotten a pointer or some other fiddly parameter to one of these objects wrong? (mine do all the time!). It's incredibly complex to create software for the Propeller!
Part of this is because of the severe limitations of the SPIN language. For example, no common object format - not even an officially defined binary format! No include files or other configuration mechanism that would allow the distribution of software other than in source form. No conditional compilation to allow multiple versions. No user libraries. Not even an official debugger! The list of omissions we expect new users to simply "deal with" is nearly endless!
Another problem is the pretence that we all seem to tolerate that SPIN is somehow "object oriented". It just isn't - at least not in the terms that any modern day "Brian" would understand. We throw the term object around, and then have to spend additional time explaining how SPIN objects are not the same as C++ or Java objects etc etc
I think these are the things that are seriously impeding the wider adoption of the Propeller - by anyone! In fact it's only the sheer brilliance of the basic hardware design that keeps us all putting up with it!
But back to the topic at hand ...
For design engineers, the biggest thing I can think of would be the opportunity to dump the SPIN interpreter - or at least make it replaceable with something more application-specific. And also not to require an external EEPROM and crystal - you can halve the component count of most small applications right there!
Oh, and as Kevin Wood just pointed out on another thread - power the bloody things from USB! I'm running out of wall adaptors and sockets!
For newbies maybe something like 12 Blocks is the right answer. Or Perhaps SPIN++
Ross.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Catalina - a FREE C compiler for the Propeller - see Catalina
Post Edited (RossH) : 7/15/2010 10:23:12 AM GMT
Cluso did open with statements about "Peripheral Blocks" and comparison tables of different Prop configurations with different Peripheral Blocks. The same Prop chip, different microcode. He does not mention "cog" once. Except his signature link to ZiCog[noparse]:)[/noparse]
Good engineers can easily pick up everything about the Prop. But they might just skip over it in the catalog when browsing around for micro-controllers because its presentation is so weird and it seems not to have all those nice peripherals they are looking for.
You have a good point about the retail/hobby market and the ease of using object "blobs". However I also think that those missing features from Spin are also what makes it so easy for beginners to get into programming. As did BASIC in days gone by.
Object format? Whats that?
Linking? Hu?
Conditional compilation? What I have to figure out what all these flags mean and set them correctly?
Libraries? Oh dear.
Do you mean "object" as in Java objects or those funny file formats you mentioned?
No, a beginner, that mythical young Brian, just want to type in a couple of lines of code and have it produce a result, now!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
-temperature, -55 to 125C is good
-lowest power in a slow mode with one cog running for sleep on battery powered apps
-price, has kept coming down, I have heard others say Prop costs so much more than pic, then are surprised to hear the price has come down to $8
-it would be nice to see some min/max logic levels of pins and 'absolute maximum ratings'
-a cut sheet or spec sheet like ones for many other maker's chips such as http://www.national.com/ds/LM/LM124.pdf . Perhaps there is one, but no obvious link to it.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Eric
I guess I didn't get my point across well enough. With the things I mentioned, we could all help build something young Brian might be able to use. Without them, everybody who comes to the Prop has to start pretty much from scratch and has to learn all about these things on their own - and then when they know enough, they naturally begins to wonder why they weren't provided in the first place. Then some of them move to a platform that does have them!
Except, possibly for things like 12 Blocks.
Ross.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Catalina - a FREE C compiler for the Propeller - see Catalina
$8 is a lot of money! I can get a 40 MIPS 16-bit PIC24 or dsPIC for under $3 that will outperform the Propeller in quite a few applications:
www.microchip.com/wwwproducts/Devices.aspx?dDocName=en537159
It comes in a little 18-pin package.
A free C compiler is available, it has very good debug facilities, and support is excellent.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Links to other interesting threads:
· Home of the MultiBladeProps: TriBlade,·RamBlade,·SixBlade, website
· Single Board Computer:·3 Propeller ICs·and a·TriBladeProp board (ZiCog Z80 Emulator)
· Prop Tools under Development or Completed (Index)
· Emulators: CPUs Z80 etc; Micros Altair etc;· Terminals·VT100 etc; (Index) ZiCog (Z80) , MoCog (6809)·
· Prop OS: SphinxOS·, PropDos , PropCmd··· Search the Propeller forums·(uses advanced Google search)
My cruising website is: ·www.bluemagic.biz·· MultiBlade Props: www.cluso.bluemagic.biz
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
Post Edited (Leon) : 7/15/2010 2:40:42 PM GMT
Whether $8 is a lot depends on the application and volume and many other things, it is comparing apples and oranges most of the time. Outperforming the propeller in a few applications is not hard, an LED is better at producing light than a propeller and cheaper too, one of the best features of the prop is the versatility and the ease of development and that for me makes it great value for money.
Graham
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
Hong Kong, and Taiwanese manufactures are commonly looking for the cheapest solution they can for toys, and other
cheap microcontroller based gadgets to get the job done. An $8.00 chip isn't attractive in these situations when they
can buy a $3.00 (or much less in bulk). I know this isn't what we want to hear, but it bares consideration.
As my 18 year old says, "The real world sucks, dad." [noparse];)[/noparse]
@Leon, did you really have to post a link too?
OBC
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Propeller Feature Projects: PropellerPowered.com
Visit the: PROPELLERPOWERED SIG forum kindly hosted by Savage Circuits.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
Time to market could be worth that, depending on the product cost and expected life cycle. Basically there is a product life-cycle, starting with early adopters, escalating in to the majority, trailing with the late adopters. That's a bell curve, for the most part too.
If there are NO competitors at all, the product will see that nice bell curve, and all the profit that goes with that cycle. Quality issues are king here as replacements / warranty are the key to keeping margins earned in house. Also, the lowest possible BOM cost is important, so that the maximum money is obtained throughout that bell curve. One variation is to lower the BOM, through revisions during that curve. That can translate into more money, but puts quality and risk on the table, just FYI.
Edit: Also, early adopters pay the most, resulting in the highest margins, and this is what makes the time to market case. The early majority along with the majority (top 'o the curve) pay a solid amount, for the majority of the dollars earned. The late adopters are cheap o, expecting mature products that perform well, at a very low cost margin. Unless the product has a very long life, or is sold in huge volumes, not a lot of money is made here, which is why being first matters so damn much, just FYI.
The real world has niches in it, some that come to exist, some that end up being created when a new product is disruptive to established niches, or where a product combines them.
The real world has competitors in it too.
Here's another fact. If there is one product exactly, everybody picks it, or not, depending on their need / want. If there are two products, identical in every way, people will choose both products, with one or the other of them easily getting a third of a share, even when they are identical in every way. Factor out ad dollars, BOM, features, even cost, and they still will choose.
Why?
Because people like selecting products, and they will choose. It might be for bogus reasons too, like I don't like her hair, or bob has that one, and he's an ***, so I'll pick this one, and so on. Happens.
Time to market then is very, very important, because whoever gets out there first, gets dominant share, making it difficult for others to take over that niche. It can be done, of course. Cost, marketing, features (and there is that word again), and other differentiators can flip who owns that niche, but risk is high, and the effort is expensive overall.
If your product life cycle is 24 months, being 6 months late to market could mean over half your profit is never realized.
That's worth $5 bucks. Could be that a Propeller gets it out there quickly, and while the product is selling, some of those dollars are used to literally do it again with a less expensive BOM, securing the longer term dollars, locking out potentially disruptive competitors. That sounds nuts, but the truth is the dollars captured by being first are often worth more than the revisions are. If that's the case, using whatever it takes to get rapid, and solid development done is worth it's weight in gold. Now you understand why executives will buy into any scheme, process, consultant, etc... who makes a compelling case for that. Most engineers hate it too, but they are not generally the ones with the money, and so it goes.
I'm serious about that. This is one area where a Prop could very easily compete, and where a very compelling business case can be made to executives, who can and do buy into those kinds of things for significant amounts of money, because that life cycle curve, and it's implications on the overall profit from a product are so significant. The down side is that somebody has to go look at the business, perform the analysis on the time to market impact, then go and tell the executives about a solution to that. They generally won't do it on their own, although they do respond to marketing and case studies.
With that kind of scenario in play, engineers will be told to develop it NOW, get it moving, then do it better faster cheaper for rev 2, 3, etc... Happens every day.
I know the thread is about engineers and selecting products. But, what I just wrote has a very significant influence on how they pick and why they pick the way they do. In the end, those picks have to align with the business goals, or the engineering team isn't doing what the executives want them to do. Those conflicts eventually get resolved, and we all know how that goes. Most corporations operate that way, and so these things must be considered.
Now, it could be that a product life cycle is very long. In that case, time to market is less of a factor, though it still is important, because of how people pick things. It also could be that many competitors are expected. Each additional one lowers the importance of time to market because the choices people make will be diluted, lowering the value of a majority share. In that scenario, cheap *** is the primary business goal. A propeller would not compete so well, unless using it brought a significant BOM cost advantage. Could happen on higher feature count products, but this case is not so easily made in a compelling way.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Propeller Wiki: Share the coolness!
8x8 color 80 Column NTSC Text Object
Wondering how to set tile colors in the graphics_demo.spin?
Safety Tip: Life is as good as YOU think it is!
Post Edited (potatohead) : 7/15/2010 4:01:29 PM GMT
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
Post Edited (Leon) : 7/15/2010 4:05:54 PM GMT
That's a very significant question. The early adopters will pay that $20, because getting the product early matters to them for their own reasons.
Read my post again. It's entirely possible to get the product out there, at a higher cost, thinner margin, capture the early momentum, then release rev 2, on a longer development cycle, securing that momentum, capturing the thicker margins. This is done all the time, with most products seeing several revisions through out their life cycle. The reasons for those revisions generally are cost reduction, quality / warranty issues, and features over time to compete for the majority share, and or maximize the dollars found over the late majority share.
If this is done before a competitor hits the niche, it's golden. If it's done late, forget it. Might as well make the cheapest product possible, because you will be selling to a fraction of the majority, and perhaps a lot of the late adopters, where margins are thin.
Profit can double on a product that has a short time to market. That's a fact.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Propeller Wiki: Share the coolness!
8x8 color 80 Column NTSC Text Object
Wondering how to set tile colors in the graphics_demo.spin?
Safety Tip: Life is as good as YOU think it is!
Parallax has been and is an industry leader in the hobby market. Doubt me? Look at all the stamp clones.
I wish I could find the video of Jeri Ellsworth talking in detail about her experiences with the Commodore DTV.
She spells out just how cut-throat that market can be. .20 is a BIG deal when talking about going to
production with a product with a 50,000 (or higher) run.
It'll certainly be interesting how Parallax meets these challenges. I'm pretty sure they will do it with style.
OBC
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Propeller Feature Projects: PropellerPowered.com
Visit the: PROPELLERPOWERED SIG forum kindly hosted by Savage Circuits.
Yep, it could be. 100 percent spot on. But, that does not mean that it IS. That's the open door for Parallax, and frankly anybody producing a highly differentiated product.
Just so you know, I make that case damn near every day. "Could be" is where the dollars are made for those people willing and able to go and properly position their product alignment with the business needs.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Propeller Wiki: Share the coolness!
8x8 color 80 Column NTSC Text Object
Wondering how to set tile colors in the graphics_demo.spin?
Safety Tip: Life is as good as YOU think it is!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
Post Edited (Leon) : 7/15/2010 4:15:07 PM GMT
Two things need to occur here. Somebody needs to run some numbers and product scenarios to obtain likely business cases. Either that person, or somebody else needs to put those into context and go tell some management about it. If they see a plausible return on investment case, the technology will be given consideration. It's that simple.
That's the job required to get commercial adoption started for all but start ups, who might choose the Prop for any number of reasons.
Oh, and I would also add that "often" need not be a qualifier. If it's possible to do at all, then some analysis of the commercial market place would reveal a niche that's favorable. That's called your market opportunity. From there, the most likely prospects would require some study, business case made, and pitched. (that part sucks, but is necessary) Edit: BTW, that analysis would reveal how much of an effort would be required, and weighing that against the opportunity would reveal whether or not the potential to profit exists. If the niche is too small, for example, more innovation would be necessary. That's prop II, in our case.
Right now, I submit that NOBODY knows, because this exercise has not yet been done.
That niche could be small, and likely is. The commercial market is absolutely huge. A few successful adoptions would be very likely to have a very significant business impact on Parallax. No need to think in terms of dominant share, only some share and what the margins are. If that's profitable, then life is good.
It's often a mistake to compare entities of such a different scale. Their business needs and overhead are much different. A small competitor with a loyal base will do well, and often have a nice, sustainable business, given they keep their focus and their product offerings relevant. Happens every day.
Leon, you are also forgetting one key element. If the business case is made, those with the dollars will fund the risk effort, essentially asking their team to give the technology a run. (asking is a nice word, often they will be told, and given support) In most companies, those kinds of decisions are made by those funding the operation. That's what needs to be exploited here, however badly that might offend an engineers sensibilities. That's just how it is.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Propeller Wiki: Share the coolness!
8x8 color 80 Column NTSC Text Object
Wondering how to set tile colors in the graphics_demo.spin?
Safety Tip: Life is as good as YOU think it is!
Post Edited (potatohead) : 7/15/2010 4:38:22 PM GMT
-Phil
I don't know about that. I once worked on a Propeller-based project which was an ideal fit for the device. Unfortunately, we were only given about two weeks to deliver a working prototype, and all I was able to do in that time was show all the various functions running on their own cogs. They still needed integrating, and the company involved pulled the plug, costing my client a lot of money.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
If you have, your statement has some merit. If not, you are not the person to make that decision. I've done both, and I know absolutely how that works cold.
...and isn't that failure a personal metric? That isn't relevant to the greater problem at hand.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Propeller Wiki: Share the coolness!
8x8 color 80 Column NTSC Text Object
Wondering how to set tile colors in the graphics_demo.spin?
Safety Tip: Life is as good as YOU think it is!
Most everybody here are Parallax return customers, and some use the Propeller for this reason alone. The people that need to be asked are the design engineers that don't hang out here, or already use the Propeller.
However, don't get too upset if their answers sound a lot like Leon.
Frankly, his commentary here on this topic is quite valuable, for exactly the reason you posted. It's all good.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Propeller Wiki: Share the coolness!
8x8 color 80 Column NTSC Text Object
Wondering how to set tile colors in the graphics_demo.spin?
Safety Tip: Life is as good as YOU think it is!
I'm no salesman, although I'm quite good at convincing people that my solution might be the best one for the job in hand. What has selling stuff got to do with the subject of this thread?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Leon Heller
Amateur radio callsign: G1HSM
That reason is the propeller presents a risk right now. It's a risk because it's new, it's a risk because it's different.
Entertaining risk costs money. We have some people here, self-funded, able to invest their money against that risk to make a go of things with a Propeller. That's a great thing everybody would like to see more of.
In the commercial space, funding that risk more or less requires executive level buy in. They, in the end, can authorize and support that risk taking, and when they do, they empower technical people to then do what they do, with the tools, dollars, and support needed to make that risk investment. The reality is they will most often do that for a return on investment case made to them, in addition to specifications and other things.
That's why I asked. And the kinds of people that do that, are not the same kinds of people that realize the technical solutions in general. I happen to do both on a regular basis, because the niche I'm in requires all involved to have some awareness of that process, and on occasion requires the sales work, in addition to the technical work. I made the posts on this topic I did, to present some of how that works in a way that frames up the problem holistically, which I believe is necessary for any real progress on propeller adoption outside the current adoption boundaries.
You've made great points. I think they are valuable. I also think some of the angst over this topic is caused by those points contributing to the perception of the whole thing being futile. I do not believe it's futile, just an effort, and a kind of effort that most current evangelists of the propeller would find unpalatable.
Edit: There is one other reason. Having features that would encourage an engineer / designer to select a propeller is a more permissive scenario where risk has been addressed in this fashion. That's the relevance to the thread. Add up all those compelling features, and we've got a list of all the cool things Propellers do. That's why we are all here, because of those cool things. No question. Now, what makes those features matter, and how are they qualified and translated into real world adoption? That's the risk, and the business end of things. Open that case up, and more features are favorable more of the time. Don't address it, and everybody then becomes very conservative, because risk costs dollars, and that generally is not a good thing, that's all.
Cheers, and I mean no harm or disrespect to anyone.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Propeller Wiki: Share the coolness!
8x8 color 80 Column NTSC Text Object
Wondering how to set tile colors in the graphics_demo.spin?
Safety Tip: Life is as good as YOU think it is!
Post Edited (potatohead) : 7/16/2010 2:20:26 PM GMT