how's the gain difference between identical ICs? Is there any?

[Thecomedian]

Transistors can range pretty widely between various values, and I wonder if ICs do the same thing, or if changing out 1 IC for an identical model IC could change tone/sound.

[GibsonGM]

You might find a very little difference in their parameters due to manufacturing differences, but I would expect them to be VERY much the same in terms of performance!    The QA in those factories is very high...

[Thecomedian]

thanks.

[Mark Hammer]

This was discussed recently, albeit under a thread whose header I can't remember.

Given that op-amp gain is not inherent to the chip but is set by external components (the very basis for the origins of op-amps AS chips), f or all practical purposes, there is no inherent "gain" difference between op-amps.  What does vary between types, however, is the extent to which they can deliver VERY large/high gain at various frequencies.  In nearly all op-amp datasheets, you will see a graph that is labelled "open-loop gain" or "gain bandwidth product" or somethng similar.  The graph shows how much gain the chip tops out at across the spectrum.  What you'll typically see is that op-amps are limited in producing much more than a gain of 3000x or greater, and this starts to decrease once you start going above 1khz.  A "great" op-amp will hold that standard up to maybe 5khz or so, and then start petering out.

Apart from a select few circuits where the limitations of the chip are exploited (e.g., the RAT), most of the overdrive and distortion circuits we all know and love can pretty much do what they do with total gain (i.e., the gain of op-amp A multiplied by the gain of op-amp B, op-amp C, etc.) of less than 500x.  For instance, the Tube Screamer uses a max gain of 118x in its clipping stage.

So, even though yes there ARE differences in the maximum gain various op-amps can provide, for all practical purposes, we do not run into the sorts of conditions where that would make any difference all that often.

[Bill Mountain]

This was discussed recently, albeit under a thread whose header I can't remember.

Given that op-amp gain is not inherent to the chip but is set by external components (the very basis for the origins of op-amps AS chips), f or all practical purposes, there is no inherent "gain" difference between op-amps.  What does vary between types, however, is the extent to which they can deliver VERY large/high gain at various frequencies.  In nearly all op-amp datasheets, you will see a graph that is labelled "open-loop gain" or "gain bandwidth product" or somethng similar.  The graph shows how much gain the chip tops out at across the spectrum.  What you'll typically see is that op-amps are limited in producing much more than a gain of 3000x or greater, and this starts to decrease once you start going above 1khz.  A "great" op-amp will hold that standard up to maybe 5khz or so, and then start petering out.

Apart from a select few circuits where the limitations of the chip are exploited (e.g., the RAT), most of the overdrive and distortion circuits we all know and love can pretty much do what they do with total gain (i.e., the gain of op-amp A multiplied by the gain of op-amp B, op-amp C, etc.) of less than 500x.  For instance, the Tube Screamer uses a max gain of 118x in its clipping stage.

So, even though yes there ARE differences in the maximum gain various op-amps can provide, for all practical purposes, we do not run into the sorts of conditions where that would make any difference all that often.

The only thing I could even think to add to this are that some chips have different biasing requirements.  If you use large biasing resistors the lack of current can have a negative effect on your chip.  For example:  I used an NTE LM308 work alike that sounded really splatty.  I checked the voltages and the Vbias was 2 volts.  I swapped the biasing network for much smaller resistors and my Vbias went back up to 4.5 volts.

[R.G.]

What Mark and Bill said, plus one additional viewpoint.

The whole point of operational amplifiers is that whatever is inside the amplifier is intended NOT to matter. The name "operational" was take to imply that the amplifier had no character of its own, only that which was conferred on it by the parts in the the rest of the circuit.

This can never be perfectly true, but modern opamps come very close. The differences between one opamp and another usually only show up on the edges of its operations: highest and lowest frequency, highest and lowest voltage, being pushed to nearly the limits of its power supply, noise content with very small signals, etc. - all the side effects and "edges" of where it works. The internal gain varies from chip to chip within the same type number, and across type numbers. But the wonders of negative feedback make the differences difficult to even detect in normal operation.

Finding the perfect opamp was almost a grail search among EEs in the 70s and 80s, and pretty much IS a grail quest among effect DIYers today. But for most applications at medium frequencies, voltages and signal excursions, they're interchangeable. Of course, effects hackers are all about the "edges" so there's a thriving gossip industry about which is better - Tweedledum or Tweedledee.

[Bill Mountain]

Oh it's definitely the NOS JRCTweedledees.

[GibsonGM]

You've seen the NOS Tweedledees???  I've been trying to find them for MONTHS, man!!!   )