| Author |
Message |
pbs681
Joined: 19 Aug 2004
Posts: 154
Helped: 11
|
 25 Sep 2008 4:51 ur opinion on technology scaling for Analog design |
|
|
|
Hi All,
I think technology scaling give more disadvantages to the analog design area. Take for example, the headroom is less, so we need to increase the size... also, the intrinsic gain is lower as we scale the tech... what do you think guys???
thanks
|
|
| Back to top |
|
 |
yxo
Joined: 13 Jul 2007
Posts: 118
Helped: 8
Location: Russia, Moscow
|
| 25 Sep 2008 10:44 ur opinion on technology scaling for Analog design |
|
|
|
Without any doubt, the scaling ads a huge amount of problems. With the scaling the speed increases only, instead we get a lot of other problems such as gain, mismatches, transconductance 
|
|
| Back to top |
|
|
fanshuo
Joined: 21 Jun 2007
Posts: 303
Helped: 5
Location: Netherlands
|
| 25 Sep 2008 13:02 ur opinion on technology scaling for Analog design |
|
|
|
pros and cons
As a designer,your only choice is to live with it and try to take advantage of it
|
|
| Back to top |
|
|
pbs681
Joined: 19 Aug 2004
Posts: 154
Helped: 11
|
| 25 Sep 2008 13:06 Re: ur opinion on technology scaling for Analog design |
|
|
|
| fanshuo wrote: |
pros and cons
As a designer,your only choice is to live with it and try to take advantage of it |
yes, it has pros and cons, but i think it has more cons than pros....
|
|
| Back to top |
|
|
JoannesPaulus
Joined: 19 Mar 2008
Posts: 58
Helped: 8
Location: POLAND
|
| 25 Sep 2008 17:26 ur opinion on technology scaling for Analog design |
|
|
|
| A very important pro is that you can use digital calibration to fix the analog problems since a few thousand gates are not going to change your size by much anymore...
|
|
| Back to top |
|
|
rfsystem
Joined: 25 Feb 2002
Posts: 797
Helped: 85
|
| 25 Sep 2008 21:23 Re: ur opinion on technology scaling for Analog design |
|
|
|
Voltage reduction and less analog features are the most serious issue:
Take a analog filter application, still more energy efficient than ADC+DSP:
Filter caps are designed for SNR. Cap size goes by C~1/VDD^2. Driving current of the filter stages is I~C. Power is then P~1/VDD
250nm 3.3V 1.00*Ibias 1.00*P
130nm 1.5V 4.84*Ibias 2.20*P
65nm 0.9V 13.44*Ibias 3.67*P
That is if there is no speed limitation. If speed limitation in some circuits you can trade more power for this filter. Or finally decide to use multichip solutions. I see it as a very popular trend to use multitech chipsets.
I remember ther was an article than depending on the specific analog function there is performance max tech, a power min tech and a cost min tech.
So depend on the application and usage what is the way to go.
|
|
| Back to top |
|
|
rf_ray
Joined: 01 Aug 2004
Posts: 25
|
| 26 Sep 2008 8:40 ur opinion on technology scaling for Analog design |
|
|
|
| Mostly cons..Some designer rarely use minimum length even it's available.
|
|
| Back to top |
|
|
