I wonder if any of the Prop II customers who like to plan ahead would be interested in purchasing an FPGA development board loaded with the Prop II beta design in the future. I don't think the A/D could be put in the FPGA as it is proven out in the die, but other features could be provided in the FPGA.
FPGAs are not cheap, unfortunately, but for those who've got commercial applications that demand the Prop II design this could represent a time saver in the design phase.
Cost? No idea. Probably $500-$2500 unfortunately. Not only is the manufacturing process more complex with BGA packages, the FPGAs are also expensive. It would only be loaded on Parallax-provided designs, too. Such a solution isn't for everybody, but might make a lot of sense for developers who would benefit from a six to nine-month head start on the actual device.
Ken Gracey
Post Edited (Ken Gracey (Parallax)) : 8/25/2008 11:41:04 PM GMT
Yes... but don't expect much feedback from me... Everything I know about FPGA comes from Nuts and Volts. I know I have to understand it eventually... and there wouldn't be better place to start...
But... unless you set up a leasing plan... $2500 is my price point. that's two zeros... not three.
Before I started with the Prop (a year ago now) I was leaning heavily toward FPGA development (mostly because it could directly drive VGA). Now that I think about it, if I had an FPGA in my hands, I might be tempted to just do everything with that...
When I discovered the Propeller I was looking for something which I could use to fulfil an on-going ambition to build a CPU. The choices seemed to be TTL ( too much work ), any old micro ( too boring ) and FPGA ( far too expensive, even for the kid's stuff, and a steep learning curve ) and then I found the Propeller. What really swung it for me was the TV output plus multi-core implemented in a manner I'd never seen so simply done.
With hindsight I made the best decision, I'm creating CPU's via Virtual Machines and that's not much different to microcoding, just easier and quicker. I'm really looking forward to the Propeller II because with 160MHz, 16 Cogs I make that around 2,500 MIPS !
What a shame it won't have interrupts [noparse]:)[/noparse] [noparse]:)[/noparse] [noparse]:)[/noparse] [noparse]:)[/noparse]
For Prop II spin interpreter, things would really speed up if SPIN could read most of the special registers directly instead of pushing it into the stack and then popping it back just to read it's value.
I want a Parallax designed FPGA board. I'm sure it would be totally awesome.
BUT
1) This has to be general purpose, not just PropII loaded, to justify the expenditure.
2) $300 max. For that we can already get an FPGA board with ARM cpu, 64Mbits SDRAM and 2 ethernet 10BASE-T from, for example, www.knjn.com/ShopBoards_USB2.html
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
Ken, are you serious? I will buy an FPGA based propeller·development system regardless of price, so sign me up.·I am into FPGA's (see this post··http://forums.parallax.com/showthread.php?p=673219·) and would love to integrate·a CPU I know and can write programs for in my designs.
Heater, the FPGA's·on the KNJN boards are relatively low end Cyclone·devices EP1C3 or EP2C5. I use·their Pluto 3 board, it is very good value for the price, but to implement multiple Cog's at speed you·need to think Stratix devices, which·are much more expenseve.
If you look back early enough (please don't) you will find one of my earliest posts begging for someone to hook a Prop up to an FPGA...
The idea of getting Prop logic as an FPGA is stunning. It would close the complete circle and allow interested people to study the Prop from the ground up... an educational opportunity that is mind boggling.
I would venture a guess that doing the same for the PROP I at a small margin to the educators and students might create a new market segment for you guys. You would know[noparse]:)[/noparse]
for example, I have some algorithms that really do belong in hardware... and there are a few instructions that I could do without... and you have your own foundry... what could be better?
Phil could make his own cameras... and I could buy them in bulk.
I got the impression that Parallax were talking about an FPGA to emulate the Prop-II to allow early adopters to get engineering time on the new chip, rather than proving a core you could incorporate into your own FPGA's or products.
I would have though that Parallax releasing any form of a FPGA source of the core of the new chip would be akin to opening it up far more than was commercially desirable.
@BradC: I think you're absolutely right - can't see even Chip releasing that much to the wild.
@Ken: Sounds like a good offer, but with a very limited market I suspect - much too expensive for the casual hobbyist like me I'm afraid.
@Coley: TWO? You're feelin' flush then (or you and Baggers have some GREAT video products just waiting for PropII's awesome power)!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cheers,
Simon
www.norfolkhelicopterclub.co.uk
You'll always have as many take-offs as landings, the trick is to be sure you can take-off again ;-) BTW: I type as I'm thinking, so please don't take any offense at my writing style
SimonL said...
TWO? You're feelin' flush then (or you and Baggers have some GREAT video products just waiting for PropII's awesome power)!
LOL
Yeah, something along those lines.....
Funnily enough we have been looking at FPGA's for a couple of projects mainly because they are beyond what we can currently do with Prop I, we were even looking at an FPGA version of PropGFX at one stage but decided it wasn't in the spirit of what we were trying to acheive.
Because our projects tend to lean more towards gaming and graphics we are hungry for more resources and speed, we have plenty of ideas that are beyond a 1 chip solution presently, that will change with the release of Prop II.
What we have created so far with PropGFX I never thought I would see from a $12 micro and that is what is the best thing for me about the Propeller, it inspires you to go that little bit further, to push the envelope that little bit more.
So no, I'm not feeling flush at all (I wish I was!) , I just know when to take an opportunity when it comes along....
Coley
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔ PropGFX Forums - The home of the Hybrid Development System and PropGFX Lite
It wont be double the silicon, a smaller feature size process will be used. As to speculation on prices, it's just that, speculation. We won't know what price·will be until we know what the actual manufacturing cost is.
You can count the number of companies that would even consider doing this... it's a pretty small number. I'm not flush either... the problem is that I see so much opportunity that trying to choose the scope and focus becomes a problem. I really like the concept of a 12 Euro box directed to people that have nothing... What thinks you?
To get back to this "I just need to narrow down what kinds of demodulation we should support"
What about something that could handle FireWire 100? There could be licensing issues but it may be able to do dv with the PropII. I assume that you are aiming at a maximum clock of 160MHz so anything faster would be out. That also means that Gigabit ethernet would also be out. Also on the subject of the counters, would it be possible to include a proper PWM mode like on the AVRs? Then the cog could be doing other stuff instead of having to constantly change the counters.
Comments
FPGAs are not cheap, unfortunately, but for those who've got commercial applications that demand the Prop II design this could represent a time saver in the design phase.
Ken Gracey
Parallax, Inc.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Paul Baker
Propeller Applications Engineer
Parallax, Inc.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Thank's Brian
·"Imagination is more important than knowledge..." ·· Albert Einstein
www.PropelX.com
Post Edited (Brian Beckius) : 8/25/2008 11:25:52 PM GMT
What price talk You on it.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Nothing is impossible, there are only different degrees of difficulty.
Sapieha
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Thank's Brian
·"Imagination is more important than knowledge..." ·· Albert Einstein
www.PropelX.com
Post Edited (Brian Beckius) : 8/25/2008 11:24:19 PM GMT
Ken Gracey
Post Edited (Ken Gracey (Parallax)) : 8/25/2008 11:41:04 PM GMT
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
--Steve
$500 but not more.
It is overcome if!
It can be open to reprogram to My own System after PropII come.
And if I can reprogram it with Yours uppdates on it!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Nothing is impossible, there are only different degrees of difficulty.
Sapieha
Post Edited (Sapieha) : 8/26/2008 12:27:10 AM GMT
Yes... but don't expect much feedback from me... Everything I know about FPGA comes from Nuts and Volts. I know I have to understand it eventually... and there wouldn't be better place to start...
But... unless you set up a leasing plan... $2500 is my price point. that's two zeros... not three.
Rich
-Phil
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
'Still some PropSTICK Kit bare PCBs left!
With hindsight I made the best decision, I'm creating CPU's via Virtual Machines and that's not much different to microcoding, just easier and quicker. I'm really looking forward to the Propeller II because with 160MHz, 16 Cogs I make that around 2,500 MIPS !
What a shame it won't have interrupts [noparse]:)[/noparse] [noparse]:)[/noparse] [noparse]:)[/noparse] [noparse]:)[/noparse]
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.fd.com.my
www.mercedes.com.my
BUT
1) This has to be general purpose, not just PropII loaded, to justify the expenditure.
2) $300 max. For that we can already get an FPGA board with ARM cpu, 64Mbits SDRAM and 2 ethernet 10BASE-T from, for example, www.knjn.com/ShopBoards_USB2.html
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
For me, the past is not over yet.
Heater, the FPGA's·on the KNJN boards are relatively low end Cyclone·devices EP1C3 or EP2C5. I use·their Pluto 3 board, it is very good value for the price, but to implement multiple Cog's at speed you·need to think Stratix devices, which·are much more expenseve.
Nico Hattink··
That is a great offer, we will have two when they are available, we areleady have lots of plans for Prop II.
Regards,
Coley
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
PropGFX Forums - The home of the Hybrid Development System and PropGFX Lite
If you look back early enough (please don't) you will find one of my earliest posts begging for someone to hook a Prop up to an FPGA...
The idea of getting Prop logic as an FPGA is stunning. It would close the complete circle and allow interested people to study the Prop from the ground up... an educational opportunity that is mind boggling.
I would venture a guess that doing the same for the PROP I at a small margin to the educators and students might create a new market segment for you guys. You would know[noparse]:)[/noparse]
Phil could make his own cameras... and I could buy them in bulk.
I would have though that Parallax releasing any form of a FPGA source of the core of the new chip would be akin to opening it up far more than was commercially desirable.
Am I wrong?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Pull my finger!
@Ken: Sounds like a good offer, but with a very limited market I suspect - much too expensive for the casual hobbyist like me I'm afraid.
@Coley: TWO? You're feelin' flush then (or you and Baggers have some GREAT video products just waiting for PropII's awesome power)!
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cheers,
Simon
www.norfolkhelicopterclub.co.uk
You'll always have as many take-offs as landings, the trick is to be sure you can take-off again ;-)
BTW: I type as I'm thinking, so please don't take any offense at my writing style
LOL
Yeah, something along those lines.....
Funnily enough we have been looking at FPGA's for a couple of projects mainly because they are beyond what we can currently do with Prop I, we were even looking at an FPGA version of PropGFX at one stage but decided it wasn't in the spirit of what we were trying to acheive.
Because our projects tend to lean more towards gaming and graphics we are hungry for more resources and speed, we have plenty of ideas that are beyond a 1 chip solution presently, that will change with the release of Prop II.
What we have created so far with PropGFX I never thought I would see from a $12 micro and that is what is the best thing for me about the Propeller, it inspires you to go that little bit further, to push the envelope that little bit more.
So no, I'm not feeling flush at all (I wish I was!) , I just know when to take an opportunity when it comes along....
Coley
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
PropGFX Forums - The home of the Hybrid Development System and PropGFX Lite
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Paul Baker
Propeller Applications Engineer
Parallax, Inc.
You can count the number of companies that would even consider doing this... it's a pretty small number. I'm not flush either... the problem is that I see so much opportunity that trying to choose the scope and focus becomes a problem. I really like the concept of a 12 Euro box directed to people that have nothing... What thinks you?
Rich
Just wondering because I think It had a sink/fan in
the pictures from the Expo.
-Phil
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
'Still some PropSTICK Kit bare PCBs left!
What about something that could handle FireWire 100? There could be licensing issues but it may be able to do dv with the PropII. I assume that you are aiming at a maximum clock of 160MHz so anything faster would be out. That also means that Gigabit ethernet would also be out. Also on the subject of the counters, would it be possible to include a proper PWM mode like on the AVRs? Then the cog could be doing other stuff instead of having to constantly change the counters.