First of all, ECL uses bipolar transistors.
Second, it is a non-saturated logic family, that is, the transistors
stay all the time in the linear region.
And finally, current is flowing all the time across some transistor, it
is only diverted from one branch to the other as the output changes state.

I don't think there is anything transcendental about the dye geometry
because ECL is a very old family, well before the recent ultra-high
speed CMOS technology.

Isaac


Em 20/08/2013 18:58, Yigit Turgut escreveu:
> Potato products seem really interesting and not just because I like potat=
o,
> they have nice specs (<1.5ns prop. delay) indeed. I will check them in
> detail. It's funny (in a positive and sympathetic way) that they only sel=
l
> over ebay. Thanks David.
>
> Prop. delay of MC10EP08 is quite impressive, I wonder which litho/wafer
> technology is used in this device. I will not have that much of a
> capacitive load thus I assume delay should stay in pS range. They provide
> IBIS models as well, cool. Thanks Kerry.
>
>
> On Tue, Aug 20, 2013 at 11:11 PM, Kerry Wentworth <
> kwentworth@skunkworksnh.com> wrote:
>
>> It depends on what you are looking for / trying to do.
>>
>> For 5V logic, an NC7SZ86 is as fast as anything I know of, at 2.8nS.
>> ECL has the MC10EP08, with a propagation of 250pS
>>
>> Kerry
>>
>>
>>
>> Yigit Turgut wrote:
>>> Hi all,
>>>
>>> What is the fastest XOR gate currently available that can be purchased =
?
>> I
>>> might be interested in other approaches (rather than IC, if not that
>> fast)
>>> that will do the switching (XOR) very fast.
>>>
>>> I wanted ask for experienced users of the community before going for
>>> hunting.
>>>
>>> Cheers
>>> y.
>>>
>> --
>> http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
>> View/change your membership options at
>> http://mailman.mit.edu/mailman/listinfo/piclist
>>

--=20
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.