It's a good question. The answer requires making some very broad assumptions which is why it was not included in the official report.
Here is a very rough pass at an answer, but please take it with a giant pinch of salt due to the guesswork involved, and lack of rigor in checking over this analysis.
Firstly for storage, the survey only asked for range of capacity so we will make the following assumption of what the storage capacity is for each studio in a given range:
Survey Response | Assumed Capacity (PB) |
500 TB or less | 0.25 |
501 TB - 1 PB | 0.75 |
1 PB - 3 PB | 2 |
3 PB - 8 PB | 5 |
8 PB or more | 10 |
Given those assumptions, this chart shows the average storage capacity per artist workstation for each bracket of studio headcount:
Now for peak render farm capacity, we'll make these assumptions:
Survey Response | Assumed Peak Render Farm Capacity (cores) |
5,000 cores or less | 2,500 |
5,001 - 12,000 cores | 8,000 |
12,000 - 25,000 cores | 18,000 |
25,000 - 50,000 cores | 37,000 |
50,000 - 100,000 cores | 75,000 |
100,000 - 250,000 cores | 170,000 |
250,000 cores or more | 280,000 |
Which then results in this graph of average peak render farm capacity per artist workstation for each bracket of studio headcount:
The shape of these results shouldn't come as a surprise - economies of scale are a real thing and apply to studios as much as anything else. The radical difference in capacity from smallest to largest studios does look startling, but I suggest the assumptions made are probably not representative of actual studio capacities so the effect is exaggerated.
Hope that helps...
Nick