Only 2nd lab in Digital System Design

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

darthsidious

Senior member
Jul 13, 2005
481
0
71
Originally posted by: chorb
dont forget your crosstalk, series terminations, pull ups/downs and ground loops.

I'm prefer digital design, but unfortunately I'm stuck doing analog for the time being

Blasphemy! I kid, I kid . Of course as an analog IC designer, I have to support my side of the fence.

But really, it's two different levels of complexity. On a per-transistor level, sure digital is easier. But you're also expected to do more complex things with it, and the layer of abstraction digital design offers is critical for that purpose. At the end of the day, it ends up being an issue of what you're more comfortable working with/are better at - system level abstraction or detail.
 

Special K

Diamond Member
Jun 18, 2000
7,098
0
76
Originally posted by: Gibson486
Originally posted by: njmodi
Originally posted by: Gibson486
I hate digital logic. Never like dit, never will. Analog for the win!!!!

Digital is so much easier than analog -> 1 and 0 :) only 2 things to learn.

sure...you tell yourself that.

Start doing real design, then you take into account thresholds, line noise, signal noise....


in the end, analog wins;)

I just prefer designing around metrics such as speed, power, area, IPC, etc. instead of weird analog concepts like gain, frequency response, input resistance, CMRR etc. ;)

 

Gibson486

Lifer
Aug 9, 2000
18,378
2
0
Originally posted by: Special K
Originally posted by: Gibson486
Originally posted by: njmodi
Originally posted by: Gibson486
I hate digital logic. Never like dit, never will. Analog for the win!!!!

Digital is so much easier than analog -> 1 and 0 :) only 2 things to learn.

sure...you tell yourself that.

Start doing real design, then you take into account thresholds, line noise, signal noise....


in the end, analog wins;)

I just prefer designing around metrics such as speed, power, area, IPC, etc. instead of weird analog concepts like gain, frequency response, input resistance, CMRR etc. ;)


that's you;)

<----brought up in audio industry.

 

BrownTown

Diamond Member
Dec 1, 2005
5,314
1
0
I'll go ahead and mostly agree with you that laying out a circuit you already 100% understand is pretty unescecarry, but I can understand where people are comming from wanted you to do it, actually getting your hands on the parts makes it more "real" then just doing it on paper or a computer simulation. That said it gets excessive at some point, and for my digital logic the final lab is just silly in that there is 20 chips you have to wire up and its all just a follow the directions and pray for the best sort of lab. Debugging hundreds of wirewrapped wires is more or less a crap shoot, if it don't work the first time you either start from scratch or give up.
 

chorb

Golden Member
Oct 7, 2005
1,272
0
0
Originally posted by: Gibson486
Originally posted by: Special K
Originally posted by: Gibson486
Originally posted by: njmodi
Originally posted by: Gibson486
I hate digital logic. Never like dit, never will. Analog for the win!!!!

Digital is so much easier than analog -> 1 and 0 :) only 2 things to learn.

sure...you tell yourself that.

Start doing real design, then you take into account thresholds, line noise, signal noise....


in the end, analog wins;)

I just prefer designing around metrics such as speed, power, area, IPC, etc. instead of weird analog concepts like gain, frequency response, input resistance, CMRR etc. ;)


that's you;)

<----brought up in audio industry.

I work in the audio industry, designing digital circuitry :Q
 

CraKaJaX

Lifer
Dec 26, 2004
11,905
148
101
Originally posted by: njmodi
Originally posted by: Gibson486
I hate digital logic. Never like dit, never will. Analog for the win!!!!

Digital is so much easier than analog -> 1 and 0 :) only 2 things to learn.

LOL.

I'm taking a Digital Design class this semester with the final coming up on Tuesday. Yesterday our professor, talking about the final, said something along the lines of "There are people in this room right now, who if they get a 25/25 on the final, they still fail. I do not see how you can dig yourself that deep of a hole. All you really need to know is 1's and 0's..." :laugh:
 

Pheran

Diamond Member
Apr 26, 2001
5,740
35
91
I don't know why you are complaining, I loved those classes. In fact, probably the best class I had at the university was a senior-level one where we designed and built a whole computer out of chips (CPU was an 80186). We also had to write an OS for it (really just a primitive monitor) in assembly language.
 

TuxDave

Lifer
Oct 8, 2002
10,571
3
71
Originally posted by: chorb
Originally posted by: Gibson486

that's you;)

<----brought up in audio industry.

I work in the audio industry, designing digital circuitry :Q

Am I guessing what we're looking at is "digitally assisted analog components". Complex analog circuits with high gain, high linearity etc scale craptastically as we continue scaling to smaller process nodes.

Lower voltage = teh suck for analog design.
Lower transistor gain = teh suck for analog design.

To compensate we start making larger transistors and complex power hungry analog circuits. To current trend is to move towards simplier, crappier analog designs with a ton of digital compensation. Digital power and area are so small compared to analog that it's essentially "free".
 

A5

Diamond Member
Jun 9, 2000
4,902
5
81
I had to make an 8bit ADC that looked even worse than that. It was so bad that I was running out of wires :p

Fortunately I'm about to finish my last circuits lab, but I'm taking an audio engineering class in the fall that will probably make me use it all again :p
 

Gibson486

Lifer
Aug 9, 2000
18,378
2
0
Originally posted by: TuxDave
Originally posted by: chorb
Originally posted by: Gibson486

that's you;)

<----brought up in audio industry.

I work in the audio industry, designing digital circuitry :Q

Am I guessing what we're looking at is "digitally assisted analog components". Complex analog circuits with high gain, high linearity etc scale craptastically as we continue scaling to smaller process nodes.

Lower voltage = teh suck for analog design.
Lower transistor gain = teh suck for analog design.

To compensate we start making larger transistors and complex power hungry analog circuits. To current trend is to move towards simplier, crappier analog designs with a ton of digital compensation. Digital power and area are so small compared to analog that it's essentially "free".


my guess is designing the actual chip itself.