diff --git a/README.md b/README.md
index 4ce9430..9f30892 100644
--- a/README.md
+++ b/README.md
@@ -75,7 +75,7 @@ trade-offs.  This project is almost entirely limited by CPU, for
 encoding and decoding audio, which means there's no reason to get an
 instance with large amounts of memory.
 
-If you want to support up to about 40 users, any single core server
+If you want to support up to about 60 users, any single core server
 should be fine.  The public instance is running on Amazon Lightsail,
 With their smallest server (512 MB RAM, 1 vCPU, 20 GB SSD,
 $3.50/month).
@@ -127,7 +127,7 @@ sudo python3 -mpip install -r requirements.txt
 
 ### Simple Configuration
 
-Handles up to ~40 users.
+Handles up to ~60users.
 
 In /etc/systemd/system/ create `uwsgi-echo-01.service` as:
 
@@ -158,7 +158,7 @@ location /api {
 
 ### Sharded Configuration
 
-Handles up to ~1000 users, at ~40/core.  The instructions below assume
+Handles up to ~1000 users, at ~60/core.  The instructions below assume
 you are using a 12 core machine: one core for nginx, one core for
 bucket brigade, and ten cores for the shards.
 
diff --git a/server.py b/server.py
index 922e099..8c8dedb 100755
--- a/server.py
+++ b/server.py
@@ -40,7 +40,10 @@
 # best-effort basis by the client.  If many people are calibrating at
 # the same time this will be exceeded, because we only check before
 # calibration.
-MAX_USERS = 22  # XXX needs tuning
+#
+# In stress testing, the server seems to do fine with 61 users, but
+# the video call might change that (stress test includes no video).
+MAX_USERS = 35 # XXX needs tuning
 
 try:
     # Grab these on startup, when they are very very likely to be the actual