> The Lattice XO2's might also be a good fit, non-volatile (no boot rom needed).
>
> The TQ100 pkg covers 256-2000 4-LUTs ( USD $4-$11 qty 1 @ DigiKey )
> The TQ144 pkg covers 640-7000 4-LUTs ( USD $7-$14 qty 1 @ DigiKey )
> http://www.latticesemi.com/products/cpld/machxo2/index.cfm
>
> Less block RAM & no DSP's compared to similar normalized LUT-count S6 parts;
> Lattice's XP2 family includes DSP blocks in a TQ144 pkg.
>
> Another thing I like about the Lattice parts is that their free
> Diamond tool includes the OEM version of Synplify.
Looks interesting. I don't know Synplify, but a OEM version of Aldec
Active-HDL is integrated, too, which I've used sometime ago and which
was very good for debugging. And the 'C' parts need only one 3.3 V
supply voltage for core and IO, so I don't need another voltage
regulator (and power supply ramp rates requirements are very easy to
fulfil: 0.01 - 100 mV/μs).
I've installed the Diamond IDE and it is very similar to Xilinx ISE. My
current design, but with a 1 MB external SRAM instead of 128 kB, fits in
125 LUTs and the VHDL file compiled without changes (just some more
warnings about unused pins).
--
Frank Buss, http://www.frank-buss.de
electronics and more: http://www.youtube.com/user/frankbuss
Reply by Brian Davis●August 4, 20122012-08-04
Frank Buss wrote:
>
> Any device which is cheaper than a Spartan 6 (I need at least 144 pins)?
> I get the XC6SLX4-2TQG144C for EUR 9.89 from Digikey. Looks like the
> cheapest Cyclone I with TQFP 144 costs EUR 9.96 at Altera, and it is
> less powerful than the Spartan 6.
>
The Lattice XO2's might also be a good fit, non-volatile (no boot rom needed).
The TQ100 pkg covers 256-2000 4-LUTs ( USD $4-$11 qty 1 @ DigiKey )
The TQ144 pkg covers 640-7000 4-LUTs ( USD $7-$14 qty 1 @ DigiKey )
http://www.latticesemi.com/products/cpld/machxo2/index.cfm
Less block RAM & no DSP's compared to similar normalized LUT-count S6 parts;
Lattice's XP2 family includes DSP blocks in a TQ144 pkg.
Another thing I like about the Lattice parts is that their free
Diamond tool includes the OEM version of Synplify.
Brian
Reply by Frank Buss●August 3, 20122012-08-03
scrts wrote:
>> Thanks, this looks good. Just 2 EUR more expensive at Digikey than the
>> Spartan 3, but more features. I think I'll use a 6SLX4.
>
> Maybe You'd like to take Altera parts into consideration?
Any device which is cheaper than a Spartan 6 (I need at least 144 pins)?
I get the XC6SLX4-2TQG144C for EUR 9.89 from Digikey. Looks like the
cheapest Cyclone I with TQFP 144 costs EUR 9.96 at Altera, and it is
less powerful than the Spartan 6.
> P.S. The project page is seems like half-lithuanian? :-)
> Thanks, this looks good. Just 2 EUR more expensive at Digikey than the
> Spartan 3, but more features. I think I'll use a 6SLX4.
Maybe You'd like to take Altera parts into consideration?
P.S. The project page is seems like half-lithuanian? :-)
Reply by maxascent●August 2, 20122012-08-02
I would forget about using an Artix until next year as I cant see anyone
having them until then.
Jon
---------------------------------------
Posted through http://www.FPGARelated.com
Reply by Frank Buss●August 2, 20122012-08-02
Martin Thompson wrote:
>
> The 7A100 is *massive* though (some smaller Artix devices got dropped last year IIRC):
>
> * ~100K "logic cells" (63k 6-LUTs, 127k FFs), 240 DSP blocks, nearly 5Mbit Block RAM!
>
> for a small cheap FPGA you still want Spartan 6 - smallest is 6SLX4:
>
> * ~4K "logic cells" (2400 6-LUTs, 4800 FFs), 8 DSP blocks, 216Kb Blockram.
> The 7A100 is *massive* though (some smaller Artix devices got dropped last year IIRC):
>
>
Does anyone (without NDA) knows why the smaller devices got dropped, if there will some smaller devices reappear, and what the costs for "normal" quantities of the 7A100 are?
Regards,
Thomas
Reply by Martin Thompson●August 2, 20122012-08-02
Frank Buss <fb@frank-buss.de> writes:
> Gabor wrote:
>>
>> OK, it took some digging, but I found the press release:
>>
>> http://press.xilinx.com/phoenix.zhtml?c=212763&p=irol-newsArticle&ID=1715211&highlight=
>
> Thanks. This was from July 17, 2012: "Xilinx, Inc. (NASDAQ: XLNX) today
> announced first shipments of its Artix�-7 Field Programmable Gate Array
> (FPGA) family."
>
> Well, let's take pot luck when they will really ship it to end customers :-)
>
>> Apparently they are starting with the XC7A100T (smallest device).
>
> This would be good for my project. I was asking, because a CPLD is just
> by a factor of 3 of 4 too small, so any small FPGA device would work.
> But I guess the new Artix devices would be still more expensive than
> e.g. the Spartan 3A, which I hope will be produced for some more years.
The 7A100 is *massive* though (some smaller Artix devices got dropped last year IIRC):
* ~100K "logic cells" (63k 6-LUTs, 127k FFs), 240 DSP blocks, nearly 5Mbit Block RAM!
for a small cheap FPGA you still want Spartan 6 - smallest is 6SLX4:
* ~4K "logic cells" (2400 6-LUTs, 4800 FFs), 8 DSP blocks, 216Kb Blockram.
Cheers,
Martin
--
martin.j.thompson@trw.com
TRW Conekt - Consultancy in Engineering, Knowledge and Technology
http://www.conekt.co.uk/capabilities/39-electronic-hardware
Thanks. This was from July 17, 2012: "Xilinx, Inc. (NASDAQ: XLNX) today
announced first shipments of its Artix™-7 Field Programmable Gate Array
(FPGA) family."
Well, let's take pot luck when they will really ship it to end customers :-)
> Apparently they are starting with the XC7A100T (smallest device).
This would be good for my project. I was asking, because a CPLD is just
by a factor of 3 of 4 too small, so any small FPGA device would work.
But I guess the new Artix devices would be still more expensive than
e.g. the Spartan 3A, which I hope will be produced for some more years.
--
Frank Buss, http://www.frank-buss.de
electronics and more: http://www.youtube.com/user/frankbuss
Reply by Rob Gaddi●August 1, 20122012-08-01
On Wed, 1 Aug 2012 06:24:02 -0700 (PDT)
Jon <jon@beniston.com> wrote:
> They started putting PRs out about these devices out over 2 years ago.
>
> Don't believe anything Xilinx says until they are in stock.
Can you think of a semi vendor for whom that statement _isn't_ true?
--
Rob Gaddi, Highland Technology -- www.highlandtechnology.com
Email address domain is currently out of order. See above to fix.