​

Generating 4-coin tosses

Big question up front

Hey Wolfram Community,
I want to generate >10^10 4-coin tosses. E.g. 10B to 1T of these:
In[]:=
Tuples[{0,1},4]
Out[]=
{{0,0,0,0},{0,0,0,1},{0,0,1,0},{0,0,1,1},{0,1,0,0},{0,1,0,1},{0,1,1,0},{0,1,1,1},{1,0,0,0},{1,0,0,1},{1,0,1,0},{1,0,1,1},{1,1,0,0},{1,1,0,1},{1,1,1,0},{1,1,1,1}}
Is there any information difference between these two methods?
1) generate tosses using a 4x1 array of 1s & 0s
RandomInteger[1,{numberOfTosses,4}]
2) generate tosses using a random integer, 1 thru 16, each representing on of the 16 possible outcomes of a 4-coin toss
RandomInteger[{1,16},numberOfTosses]
Method #2 is much faster and my old 2013 MBP can generate 1-2 orders of magnitude more data before choking.
Would love your thoughts on this or you can just point me to some useful articles/docs if you're too busy to explain.
​
Thanks,
​
David

Background

I'm working on this quiz and my old computer is having trouble generating more than 10^8 4-coin tosses​​
Between the two methods above, #2 is materially faster
In[]:=
numberOfTosses=10000000;​​AbsoluteTiming[RandomInteger[1,{numberOfTosses,4}];]​​AbsoluteTiming[RandomInteger[{1,16},numberOfTosses];]
Out[]=
{0.233657,Null}
Out[]=
{0.062318,Null}
I am concerned that #2 encodes less information than #1, but I am out of my depth here and would love to know if that's actually true or if they can be used interchangeably.
​
a) are they equivalent from an information perspective?
b) if not, is #2 "good enough" for to answer this quiz, but this method shouldn't be used for other inquiries? (i.e. #2 contains the info needed for this quiz but not necessarily other mathy problems)
c) what analyses can I do to figure this out myself?

Solving the quiz

For anyone that is interested, here is how I solved the quiz (using data generated via method #1)
Generate Data
In[]:=
cointosses=1000000;​​dataset=RandomInteger[1,{cointosses,4}];
Count the outcomes, filter out the desired outcomes (HHHH vs HHTT), and find the ratio
In[]:=
First[#]/Last[#]&@Values[KeyTake[Counts[dataset],{{0,0,0,0},{0,0,1,1}}]]//N
Out[]=
0.998604
To save on memory/compute time I manually repeat the above until my old Mac chokes on generating dataset and/or the Counts function.