I think a lot will depend on whether it can do stuff like high-speed USB and 100 Mbit/s Ethernet in software. That might open up a lot of markets as it has done for it's main competitor.
I also wonder if the new speed limit will alter the array of high-def video modes that it supports. ('Not one of my major bullet points, but it might be for some.)
Leon
I think a lot will depend on whether it can do stuff like high-speed USB and 100 Mbit/s Ethernet in software. That might open up a lot of markets as it has done for it's main competitor.
I seriously doubt it Leon there are so many devices out there that do it in hardware. I think the prop 2 will be a product in search of a market. To many, the P8X32A is a novelty and I would seriously question what the market penetration outside of education and hobbyists is.
If I recall correctly I saw a post that says the sale of the one millionth P8X32A is coming soon, considering it was released in 2006 that's 1 million chips sold in six years (give or take).
Target the prop 2 for the imbedded market as the ability to do tasks in parallel is good feature.
I still think an upgrade for the P8X32A is a better idea and faster to market means faster ROI.
Just some random thoughts....................
Whilst many here will not agree with me I am looking at this from a different paradigm.
The advantage in doing fast USB and Ethernet in software is the low cost and high performance achievable for things like high-end audio over USB and Ethernet. The other company is doing very well with customers using their hardware and software; it's a niche market and they don't have any competition. It seems very profitable.
Granted those are good points however like you said they are niche markets and by the time the prop 2 is ready and if that is the market they wish to target then it will be starting from a long way behind. Safer to be a generalist.
Declaring "the other company" to be Parallax' chief competitor gives "the other company" far too much credit. In a world of billions of PICs, AVRs, ARMs, Arduino's, and every other flavor of embedded controller, "the other company" is currently a drop in the bucket. It's in your mind, Leon. It's all in your mind.
It's actually doing rather well. They will be in profit next year, and will probably be going for an IPO. What other company makes a multi-core embedded processor that has been accepted by professional users?
Just so everyone knows, the next Propeller chip will NOT be able to do 100Mbps ethernet, nor high-speed 480Mbps USB. Neither will it do HDMI, DisplayPort, or DVI. It will not do the new Thunderbolt protocol, either. It will, however, be able to achieve 300bps serial so that Hayes modems can be put back to work. Even for that application, though, you will need an extra chip for the +/-12V interfacing. You might be able to make a faithful Timex Sinclair emulator, with extra parts, of course.
It will, however, be able to achieve 300bps serial so that Hayes modems can be put back to work.
Oh, phew! That's a relief!
Seriously, though, are there I/O modes that you were planning to accommodate at 160 MHz that won't be possible at 100? I know HDMI was out of the picture even at 160, but I'm not sure now which parts of your above post to take with a grain of salt.
Declaring "the other company" to be Parallax' chief competitor gives "the other company" far too much credit. In a world of billions of PICs, AVRs, ARMs, Arduino's, and every other flavor of embedded controller, "the other company" is currently a drop in the bucket. It's in your mind, Leon. It's all in your mind.
From my vantage point K2 is correct. The only time I've ever heard the other company mentioned as a substitute or competitor for the Propeller is on these forums. Having worked with most all of our commercial developers, their most common alternatives they discuss are the PIC, AVR, or the SX. Once in a while we hear about the ARM, too. For hobby and education it's mostly the Arduino. The reasons the Propeller is chosen is for it's shining points (quick proto to production, multi-core design, I/O flexibility, design-in service, a more interesting development experience, reliable supply, and because the fine folks on these forums). The types of applications are usually newer or evolving technologies in small and mid-size companies with production runs between 1,000 and 100,000 units per year.
Parallax is also a drop in the bucket in the world of billions processors. Our drop size is increasing steadily, and our expectations are realistic.
One oversight that some developers make is understanding the importance of a supplier's stability and funding sources. Is there a funding source or has the company generated it's own revenue to pay for the process? Both approaches have merits, but more often than not the future supply chain is more stable with a company who grows the market and generates their own revenue.
Just so everyone knows, the next Propeller chip will NOT be able to do 100Mbps ethernet, nor high-speed 480Mbps USB. Neither will it do HDMI, DisplayPort, or DVI. It will not do the new Thunderbolt protocol, either. It will, however, be able to achieve 300bps serial so that Hayes modems can be put back to work. Even for that application, though, you will need an extra chip for the +/-12V interfacing. You might be able to make a faithful Timex Sinclair emulator, with extra parts, of course.
Granted those are good points however like you said they are niche markets and by the time the prop 2 is ready and if that is the market they wish to target then it will be starting from a long way behind. Safer to be a generalist.
Anything that is selling less volume than millions of units per year is a niche product. The mainstream MCUs, such as PIC, R8, and ARM are selling in the hundreds of millions and the PropII can never hope to even come close to that. It should focus on its strengths and the niche market established by its little brother, the PropI.
If you're still here, a more general question relating to speed: I know you've got the CORDIC stuff integrated into the hardware. For outputs up to, say, 30 MHz, how is the spectral purity affected by the processor speed (and hence, I assume, the sample rate)?
-Phil
BTW, I still wish the OP would change the title of this thread. I almost hate posting here.
It will, however, be able to achieve 300bps serial so that Hayes modems can be put back to work. Even for that application, though, you will need an extra chip for the +/-12V interfacing.
Yes! I knew there was a reason I kept my old 300 baud Hayes Smartmodem (with the nice heavy aluminum case). I'm sure I've got some LM1488s and LM1489s around too so I should be all set if they are happy with 3.3v.
Wow! They might turn a profit next year? That's fantastic! I'm pretty sure every other chip manufacturer in the world is freaking out right now.
They've only been trading for about two years. For a semiconductor manufacturer to be making a profit in such a short space of time is quite unusual. Some large companies like NXP have never made a profit.
Just so everyone knows, the next Propeller chip will NOT be able to do 100Mbps ethernet, nor high-speed 480Mbps USB. Neither will it do HDMI, DisplayPort, or DVI. It will not do the new Thunderbolt protocol, either. It will, however, be able to achieve 300bps serial so that Hayes modems can be put back to work. Even for that application, though, you will need an extra chip for the +/-12V interfacing. You might be able to make a faithful Timex Sinclair emulator, with extra parts, of course.
Thank Goodness! Finally, I will be able to keep up with the text on the screen without having to look up the details on that waitcnt command again!!
They don't really compete at present, but it will be different when the Propeller 2 arrives. The devices will be similar in terms of cost, performance and number of I/Os.
Yes! I knew there was a reason I kept my old 300 baud Hayes Smartmodem (with the nice heavy aluminum case). I'm sure I've got some LM1488s and LM1489s around too so I should be all set if they are happy with 3.3v.
But do you still have a telephone that will fit in the acoustic coupler? And you'll need an adapter to go from the 25-pin RS-232 connector to a 9-pin connector to a serial-to-USB converter. You'll probably need a gender-bender and a null-modem adapter in line with that also.
Inmos led the world with their technology 30 years ago. Lots of my transputer systems were sold to leading companies, universities and government research establishments.
But do you still have a telephone that will fit in the acoustic coupler? And you'll need an adapter to go from the 25-pin RS-232 connector to a 9-pin connector to a serial-to-USB converter. You'll probably need a gender-bender and a null-modem adapter in line with that also.
It's a direct connect modem - no coupler required, very advanced. I've got all the adapters I would need and I know where they are. Even if I didn't, I still have plenty of 9 and 25 pin plugs/jacks to make whatever I need.
For the Prop 2 interface, I would probably skip the adapters and build a custom cable - the Prop 2 could probably fit pretty comfortably in a 25-pin shell.
Just so everyone knows, the next Propeller chip..... It will, however, be able to achieve 300bps serial so that Hayes modems can be put back to work. Even for that application, though, you will need an extra chip for the +/-12V interfacing. You might be able to make a faithful Timex Sinclair emulator, with extra parts, of course.
Well, I at least expect that the PropII will be able to do the "whole" modem except the DAA/ isolation! Then we don't require the Hayes modem or the serial port at all! Therefore, no +/-12V either. (I nominate Phil to do the Bell103 code. I coded the AT command set in the mid 80s, so I could do that. - does this sound like an interesting retro project??? -not likely- )
Phil: What would you like me to change the title to???
And yet another thread that has digressed to "that" chip!!!
Something not so overtly negative maybe? How about something like, "Prop II: bigger, better, faster, more -- but by how much?" It has a more positive spin but still leaves people room to express whatever concerns they may have.
It amazes me how much several of you seem to hate or misunderstand the use of debuggers for code. Being a professional programmer, I use debuggers daily. When you are dealing with very large projects (multi-million lines of code) it's impractical to not have/use a debugger. I think it's ludicrous to say they should be avoided or that they are not needed. It's like saying you should do electronics without a multimeter and/or scope. Also, unit testing and bottom up approaches only get you so far when dealing with large complex projects. I do use unit testing and bottom up approaches a fair amount, but it's silly to think that they could replace a good debugger.
I somewhat agree. One has to understand, though, that there are various kinds of debuggers. In real-time event driven systems based on state machines, you can get very useful debugging information from a real-time spy trace QSpyl. This can be added at minimal overhead, even in fairly tiny systems (think 4kb of RAM). This trace will show -- with timestamps -- all state transitions, event broadcasts, task switches, and can include any custom data records you want to add. The datastream is robust (uses HDLC-like framing) and has very low overhead. It has been a very helpful approach in some of my designs. You can even leave it in a production system for in-the-field diagnostics.
Comments
-Phil
I seriously doubt it Leon there are so many devices out there that do it in hardware. I think the prop 2 will be a product in search of a market. To many, the P8X32A is a novelty and I would seriously question what the market penetration outside of education and hobbyists is.
If I recall correctly I saw a post that says the sale of the one millionth P8X32A is coming soon, considering it was released in 2006 that's 1 million chips sold in six years (give or take).
Target the prop 2 for the imbedded market as the ability to do tasks in parallel is good feature.
I still think an upgrade for the P8X32A is a better idea and faster to market means faster ROI.
Just some random thoughts....................
Whilst many here will not agree with me I am looking at this from a different paradigm.
Shalom.
Granted those are good points however like you said they are niche markets and by the time the prop 2 is ready and if that is the market they wish to target then it will be starting from a long way behind. Safer to be a generalist.
Declaring "the other company" to be Parallax' chief competitor gives "the other company" far too much credit. In a world of billions of PICs, AVRs, ARMs, Arduino's, and every other flavor of embedded controller, "the other company" is currently a drop in the bucket. It's in your mind, Leon. It's all in your mind.
-Phil
-Phil
I still maintain that the two companies will be competing for the same market, to some extent.
Seriously, though, are there I/O modes that you were planning to accommodate at 160 MHz that won't be possible at 100? I know HDMI was out of the picture even at 160, but I'm not sure now which parts of your above post to take with a grain of salt.
Thanks,
-Phil
From my vantage point K2 is correct. The only time I've ever heard the other company mentioned as a substitute or competitor for the Propeller is on these forums. Having worked with most all of our commercial developers, their most common alternatives they discuss are the PIC, AVR, or the SX. Once in a while we hear about the ARM, too. For hobby and education it's mostly the Arduino. The reasons the Propeller is chosen is for it's shining points (quick proto to production, multi-core design, I/O flexibility, design-in service, a more interesting development experience, reliable supply, and because the fine folks on these forums). The types of applications are usually newer or evolving technologies in small and mid-size companies with production runs between 1,000 and 100,000 units per year.
Parallax is also a drop in the bucket in the world of billions processors. Our drop size is increasing steadily, and our expectations are realistic.
One oversight that some developers make is understanding the importance of a supplier's stability and funding sources. Is there a funding source or has the company generated it's own revenue to pay for the process? Both approaches have merits, but more often than not the future supply chain is more stable with a company who grows the market and generates their own revenue.
And we have Chip.
Ken Gracey
300bps is to fast.
Anything that is selling less volume than millions of units per year is a niche product. The mainstream MCUs, such as PIC, R8, and ARM are selling in the hundreds of millions and the PropII can never hope to even come close to that. It should focus on its strengths and the niche market established by its little brother, the PropI.
If you're still here, a more general question relating to speed: I know you've got the CORDIC stuff integrated into the hardware. For outputs up to, say, 30 MHz, how is the spectral purity affected by the processor speed (and hence, I assume, the sample rate)?
-Phil
BTW, I still wish the OP would change the title of this thread. I almost hate posting here.
Yes! I knew there was a reason I kept my old 300 baud Hayes Smartmodem (with the nice heavy aluminum case). I'm sure I've got some LM1488s and LM1489s around too so I should be all set if they are happy with 3.3v.
I agree.
They've only been trading for about two years. For a semiconductor manufacturer to be making a profit in such a short space of time is quite unusual. Some large companies like NXP have never made a profit.
Who is currently funding them? Is it VC money?
Thank Goodness! Finally, I will be able to keep up with the text on the screen without having to look up the details on that waitcnt command again!!
OBC
http://en.wikipedia.org/wiki/XMOS
Ken,
They don't really compete at present, but it will be different when the Propeller 2 arrives. The devices will be similar in terms of cost, performance and number of I/Os.
NXP made a loss last year.
It's a direct connect modem - no coupler required, very advanced. I've got all the adapters I would need and I know where they are. Even if I didn't, I still have plenty of 9 and 25 pin plugs/jacks to make whatever I need.
For the Prop 2 interface, I would probably skip the adapters and build a custom cable - the Prop 2 could probably fit pretty comfortably in a 25-pin shell.
Inmos was owned by the Labour government at the time, it didn't have to make a profit.
Well, I at least expect that the PropII will be able to do the "whole" modem except the DAA/ isolation! Then we don't require the Hayes modem or the serial port at all! Therefore, no +/-12V either. (I nominate Phil to do the Bell103 code. I coded the AT command set in the mid 80s, so I could do that. - does this sound like an interesting retro project??? -not likely- )
Phil: What would you like me to change the title to???
And yet another thread that has digressed to "that" chip!!!
-Phil
OBC