Dave, the question is not trivial.  This really goes to the heart of how digital storage oscilloscopes work.  We have tools to help you with this:

http://www.autonerdz.com/java/SampleRateCalculator.html

Basically, true sample rate is a function of maximum ADC speed, memory, and capture time.   To visualize this, imagine a scope that can store 512 samples in it's memory and the ADC is capable of 25MHz.  What is your sample rate for a one second capture time?

Answer:  512 samples per second.  It makes no difference that the ADC can go much faster.  There just is no place to put the samples.  You can't overfill the memory. 

So, you either have to slow the ADC down or have some software that will decide which samples will be stored.  The true sample rate is still 512 samples per second.

So what happens with this scope if you cut the capture time down to one half second?

Answer: You just doubled your sample rate to 1024 samples per second.

Same memory, same number of samples, but spread over less time.  Therefore more samples per second.

To achieve better results you have to increase the memory so you can store more samples. 

This is the basic concept.  Obviously it can get more complex with things like shared ADC and memory and multiple channel use.  How memory is managed, and more.

When you use the automotive presets in Pico, or any presets from anyone, the sample rate setting is the preference of the person that made the preset, not how the scope can perform.  The sample rate may be deliberately turned down, or the preset may be from a much less capable model scope.  Or it may be turned up beyond the scopes' ability.  It's just what you are requesting the scope to do.  It will do it's best to comply but true sample rate will rarely match requested exactly.

I hope this was precise enough.  This is covered in way more depth than is possible here, on the Nerd 1 CD.

The Pico 3000 series automotive scopes have a 20MHz ADC and a 512,000 internal sample memory.  Both of these are shared.  You may use these numbers in the sample rate calculator to see performance at any given capture time. 

The rules get bent though, when Pico 6 goes into high speed streaming and uses the PC RAM instead of the internal memory for sample storage.  Now the jump from a one second capture time to a two second capture time results in a 12 times faster sample rate and is not affected by the number of channels in use anymore.  Now we can collect over 8 million samples on a single pass. 

This is where other scopes become a small dot in the rear view mirror.   

Tom

Its a little of topic of pico scope but is there a way to test sample rate for other scopes that do not list there buffer size?  The reason I ask is the at the shop I have access to the otc scope and use it once in awhile it would be kind of nice to know its exact limitations.  Of course with the images it puts out the performance seems pretty poor.

One other question.  The otc scope has a record feature that records 25 (screens?)  when the data is reviewed it possible to just scroll through the entire record does this mean its a seamless record like or is there a way to test this?

Here is a screenshot

Thanks Sam