which is Low power FPGA?

Hi,

Please advice me on low power FPGA available in the market. which is the low power consumption FPGA among Xilinx - Spartan3L, Altera - CycloneII, Actel - proASIC3 and Lattice. Is there any review available on this?

According to me proASIC3 is with low power, but its a flash based FPGA. Is there any disadvantages with Flash based FPGAs?

Thanks in advance.

Best regards, HimaSSK.

Reply to
himassk
Loading thread data ...

I think that flash based FPGA has advantages only :) But they could have smaller capacity then other chips. First you need to look in datasheets on quiscient power for chips you interested in.

Spartan3L could have 100-200mW.

regards

Jerzy Gbur

Reply to
jerzy.gbur

schrieb im Newsbeitrag news: snipped-for-privacy@g44g2000cwa.googlegroups.com...

read the datasheet for 3L !

the 3L has NO POWER advantages over s-3 while operationg! NONE! exactly the same power consumption as S3

3L only has additional hibernate mode, but that means the FPGA is not configured during low power hibernation.

Antti

Reply to
Antti Lukats

I could nt make out the best low power FPGA among the Spartan3, Cyclone II and Lattice FPGAs.

Regards, Himassk

Reply to
himassk

You need to work it out for your application: using the vendors mA(Static) and mA/MHz numbers, and also determine if Typical, or Worst case matters most to you. Actual measurements will probably be needed as reality checks.

Not all these numbers are clearly documented, and with everyone claiming to be lowest power, your task will not be easy.

In particular, Look for what they do NOT say.

example: I see lattice make big claims for 4000Z static Icc over coolrunner 2, but are very quiet on mA/MHz. Perhaps that number does not stack up as well ?

-jg

Reply to
Jim Granville

When you are interested in low power, you must analyze static and dynamic power separately. Static power (when the chip is powered up, but not being clocked) used to be very low, but is significant in the newer smaller-geometry technology. Static = leakage current depends strongly on temperature. Make sure you look at both 25 degree and 85 degree values. Some manufacturers conveniently underreport by mentioning 25 degree values only :-(

Dynamic power comes from the charging and discharging of capacitances, and depends therefore on the clock rate of all the various nodes inside the chip and the I/O. It is proportional to clock frequency, and increases with the square of the supply coltage (the current increases linearily, the power obviously with the square of Vcc). Newer technology reduces internal capacitances, thus use less power at the same functionality and speed. But functionality as well as clock rate often increases, and power might therefore go up anyhow.

Peter Alfke, Xilinx Applications

Reply to
Peter Alfke

none of them are really, however you can run the later ones at a considerably reduced voltage to achieve dramatic power savings at the cost of slower operation.

--
--Ray Andraka, P.E.
President, the Andraka Consulting Group, Inc.
 Click to see the full signature
Reply to
Ray Andraka

OK, you right, you don't have to scream :) But himassk asked about S3L, and I gave him answer about it.

regards. Jerzy Gbur

Reply to
jerzy.gbur

If you are looking for low power static, you can't beat the XP or XO families from Lattice. They have a power down pin that reduces the power to ~ 100 micro amps. Of course the I/O tristate and such, but since these devices are flash + sram, the part boots up in less than a mili second. If Dynamic power is important and you can fit into a cpld, look at the CR2 or Mach 4000Z.

Reply to
Teo

Jim,

If I'm not mistaken, the datasheet of the 4000Z provides both static power, and a graph showing the dynamic power requirements. There is even a complete technical note about the power coefficients.

On the FPGA side, the Power Calculator can be used for static power, and gives a very good idea of the requirements when a design is loaded. Unfortunately this is only true if a full timing simulation is done and the test vectors are fed back to the power calculator. If you don't simulate, then you need to give an activity factor.

Luc

Reply to
luc

Yes, but I was refering to the Web hoopla, [ There are NO relative claims in the Data sheets ] see

formatting link

and not a squeak on mA/MHz ?

[Aug-27-04 is also getting a tad long in the tooth ? ]

ISTR a quick compare did confirm that mA/MHz number was not so great....

-jg

Reply to
Jim Granville

Take another look at that web page. In the "Family Member Selector Guide" table the 6th column is labeled "Typical Standby Current (A)". I'm gueesing they meant uA, not Amps!

-Dave

Reply to
Dave Pollum

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.