diff --git a/CONFIGURATION.md b/CONFIGURATION.md
index 8b747e33..f75fed21 100644
--- a/CONFIGURATION.md
+++ b/CONFIGURATION.md
@@ -135,9 +135,11 @@ The default behaviour is to advertise minimum, preferred and maximum buffer
 sizes of 1 ms, 20 ms and 1 s, respectively. However, in practice the minimum
 buffer size will be advertised as 10 ms as long as the (default) DirectSound
 backend is used and the input device isn't disabled; this is to work around a
-[known issue](https://github.com/dechamps/FlexASIO/issues/50) with DirectSound
-and small input buffer sizes. The resulting sizes in samples are computed based
-on whatever sample rate the driver is set to when the application enquires.
+[known issue][issue50] with DirectSound and small input buffer sizes. The
+resulting sizes in samples are computed based on whatever sample rate the driver
+is set to when the application enquires. In addition, by default, FlexASIO will
+not advertise any buffer sizes smaller than 32 samples as that tends to [confuse
+some applications][issue88].
 
 ### `[input]` and `[output]` sections
 
@@ -366,6 +368,8 @@ automatic conversion mechanism as the one this option controls.)
 [device]: #option-device
 [GUI]: https://en.wikipedia.org/wiki/Graphical_user_interface
 [INI files]: https://en.wikipedia.org/wiki/INI_file
+[issue50]: https://github.com/dechamps/FlexASIO/issues/50
+[issue88]: https://github.com/dechamps/FlexASIO/issues/88
 [logging]: README.md#logging
 [official TOML documentation]: https://github.com/toml-lang/toml#toml
 [portaudio287]: https://app.assembla.com/spaces/portaudio/tickets/287-wasapi-interprets-a-zero-suggestedlatency-in-surprising-ways
diff --git a/src/flexasio/FlexASIO/flexasio.cpp b/src/flexasio/FlexASIO/flexasio.cpp
index d54d2b8a..74c4e098 100644
--- a/src/flexasio/FlexASIO/flexasio.cpp
+++ b/src/flexasio/FlexASIO/flexasio.cpp
@@ -387,12 +387,13 @@ namespace flexasio {
 		}
 		else {
 			Log() << "Calculating default buffer size based on " << sampleRate << " Hz sample rate";
-			*minSize = long(sampleRate * (hostApi.info.type == paDirectSound && inputDevice.has_value() ?
+			// We enforce a minimum of 32 samples as applications tend to choke on extremely small buffers - see https://github.com/dechamps/FlexASIO/issues/88
+			*minSize = (std::max<long>)(32,  long(sampleRate * (hostApi.info.type == paDirectSound && inputDevice.has_value() ?
 				0.010 :  // Cap the min buffer size to 10 ms when using DirectSound with an input device to work around https://github.com/dechamps/FlexASIO/issues/50
 				0.001    // 1 ms, there's basically no chance we'll get glitch-free streaming below this
-			)); 
-			*maxSize = long(sampleRate); // 1 second, more would be silly
-			*preferredSize = long(sampleRate * 0.02); // 20 ms
+			)));
+			*maxSize = (std::max<long>)(32, long(sampleRate)); // 1 second, more would be silly
+			*preferredSize = (std::max<long>)(32, long(sampleRate * 0.02)); // 20 ms
 			*granularity = 1; // Don't care
 		}
 		Log() << "Returning: min buffer size " << *minSize << ", max buffer size " << *maxSize << ", preferred buffer size " << *preferredSize << ", granularity " << *granularity;