BeMicro CV A9 - 301.000K LE
6581
Posts: 132
I see a lot of people (including me) are using the BeMicro CV (A2) for evaluating
of different FPGA projects including the P1V. Now it seems that there will be the
BeMicroCV A9 coming up very soon:
http://parts.arrow.com/item/detail/arrow-development-tools/bemicrocva9
It has 301.000 LE (instead of the 25.000 LE the A2 has) and a lot more multiplier
units etc.
Now, the only issue I see: It seems to be completely unsupported by the Web Edition
of Altera Quartus II. There is a subscription edition, which is around 2000$ a year and
comes with an evaluation time of 30 days.
Then I don't understand what the target market of Arrow is with the A9.. I don't think
somebody would save money on an FPGA board if the software to use it costs so much.
Am I seeing it wrong? Looks like I will be cancelling my 4 x A9 preorders.
of different FPGA projects including the P1V. Now it seems that there will be the
BeMicroCV A9 coming up very soon:
http://parts.arrow.com/item/detail/arrow-development-tools/bemicrocva9
It has 301.000 LE (instead of the 25.000 LE the A2 has) and a lot more multiplier
units etc.
Now, the only issue I see: It seems to be completely unsupported by the Web Edition
of Altera Quartus II. There is a subscription edition, which is around 2000$ a year and
comes with an evaluation time of 30 days.
Then I don't understand what the target market of Arrow is with the A9.. I don't think
somebody would save money on an FPGA board if the software to use it costs so much.
Am I seeing it wrong? Looks like I will be cancelling my 4 x A9 preorders.
Comments
$3000 for A9 US$!
Ouch.
The BEMICRO CV A9 is around $50 as well!
I think for most of us the A9 is out of our league...
===Jac
It does seem a strange idea, to NOT have a Board that has the largest-Web-Edition device.
After all, they do crow about footprint compatible FPGAs so the effort needed here is quite small.
I agree, but I think that's unlikely to happen: The A9 FPGA contains an ARM core, and Altera probably has to pay license fees to ARM to integrate the compiler in their Quartus tool so they can't afford to give that away.
The DE2-115 has the biggest available Cyclone IV FPGA on it (115K LE's) and it's supported by the free version of Quartus. I think we should be pretty grateful for that :-)
Also, Cyclone V compilations are so slow (at least for the BeMicro CV), that I'm not so sure if I'd be interested in having a version that has 301K LE's. It'll probably take the better part of a day to fill that thing up :-)
===Jac
There is also the De2i-150, which has 150K LE's, I assume it works with the free quartus web edition but don't know for sure.
http://www.terasic.com.tw/cgi-bin/page/archive.pl?Language=English&CategoryNo=139&No=529&PartNo=2
50 for the BeMicro A9 version, wow. We should all get ultra proficient on BeMicro CV's first, so that we can get the maximum from our 30 days of serious experimenting
Are you sure? The cheapest A9 chip costs around 300$. How can they make a whole board for 1/6 the price?
Andy
It's actually the SE variant having the ARM-Core. The A9 which is on the BeMicro CV A9 is the E variant which has no ARM-core.
I got the price confirmation and the board will cost around 50$ (list price). I am currently in contact with the vendor, because they apparently
didn't think about the license issue. I guess there will be some custom licensing available.
By the way: I'd personally like to go with the DE2, but $555 is a bit much compared to the BeMicroCV for 38$ and CVA9 for 50$.
Then it probably would make more sense to go with multiple BeMicro CV boards.
I see the price is now
BEMICROCVA9 Arrow Development Tools $149.0000
but shows Zero Stock, so maybe they still talk with Altera ?
If the License issue persists, then a BEMICROCVA7 would be an obvious product. ?
What to do with 300k LE's and 12 MBits of RAM?
Interesting, I wonder where that places the Parallax FPGA board ?
If you think of the 1-2-3 board not as a 'hard specced board that needs to compete with the spectrum of other fpga dev boards out there', but more as a Parallax supported way to early-adopt Chip's current and future visions (while Parallax simultaneously reduce risk and flush out bugs), then its all good.
I think Chip will release builds for a small set of FPGA boards, so this would be a clear candidate.
I guess the 1-2-3 Board and BeMicro CV A9 can co-exit depending on where Parallax plan to go with HW IO Support, and when that can no long fit on an industry board like BeMicro CV A9
eg If they want to add P2 support for SDRAM of some type, then a board with that specific chip becomes an important test vehicle,
Likewise for Hyperbus, or TIs uPP connections.
The connectors on this looked interesting
http://www.ti.com/tool/LAUNCHXL-F28377S
TI seem to have used a thru hole reflow header that has a removable cap ? - gives a nice compact PCB footprint.
I have to admit, the board is an excellent price, even with $35 shipping to Oz. I am tempted
However there's a kind of fragmentation of attention if everyone's using a different target, as well as prioritizing different P1V extensions
For the new P2, I think it'd be good if we're all (or as many as possible) using the Parallax 1-2-3 board and its unique features.
Agreed. As much as I'd like to help test the upcoming P2 image with the 1-2-3 board, it is just too cost prohibitive (as a casual hobbyist). And, of course, I already have the DE0-Nano and the original BEMicro CV, so would prefer to leverage that investment. The CV A9 is very enticing, but I'm afraid it would end up like the other two: mostly collecting dust. And that is the same issue with the 1-2-3 board, only more so. Beyond the immediate use case of beta testing the P2, I just don't have another use case that makes it worth buying.
Better start working out a good use case Seairth, that'd solve it
Better start working out a good use case Seairth, that'd solve it