diff --git a/src/Microsoft.ML.Recommender/MatrixFactorizationTrainer.cs b/src/Microsoft.ML.Recommender/MatrixFactorizationTrainer.cs
index 00153171cf..ac6a89f6d8 100644
--- a/src/Microsoft.ML.Recommender/MatrixFactorizationTrainer.cs
+++ b/src/Microsoft.ML.Recommender/MatrixFactorizationTrainer.cs
@@ -513,7 +513,7 @@ private MatrixFactorizationModelParameters TrainCore(IChannel ch, RoleMappedData
 
         private SafeTrainingAndModelBuffer PrepareBuffer()
         {
-            return new SafeTrainingAndModelBuffer(_host, _fun, _k, _threads, _threads == 1 ? 1 : Math.Max(20, 2 * _threads),
+            return new SafeTrainingAndModelBuffer(_host, _fun, _k, _threads, 2 * _threads + 1,
                 _iter, _lambda, _eta, _alpha, _c, _doNmf, _quiet, copyData: false);
         }
 
diff --git a/test/Microsoft.ML.Tests/TrainerEstimators/MatrixFactorizationTests.cs b/test/Microsoft.ML.Tests/TrainerEstimators/MatrixFactorizationTests.cs
index a6c0093242..09ce334a79 100644
--- a/test/Microsoft.ML.Tests/TrainerEstimators/MatrixFactorizationTests.cs
+++ b/test/Microsoft.ML.Tests/TrainerEstimators/MatrixFactorizationTests.cs
@@ -96,10 +96,10 @@ public void MatrixFactorizationSimpleTrainAndPredict()
             // MF produce different matrices on different platforms, so check their content on Windows.
             if (RuntimeInformation.IsOSPlatform(OSPlatform.Windows))
             {
-                Assert.Equal(0.301269173622131, leftMatrix[0], 5);
-                Assert.Equal(0.558746933937073, leftMatrix[leftMatrix.Count - 1], 5);
-                Assert.Equal(0.27028301358223, rightMatrix[0], 5);
-                Assert.Equal(0.390790820121765, rightMatrix[rightMatrix.Count - 1], 5);
+                Assert.Equal(0.290507137775421, leftMatrix[0], 5);
+                Assert.Equal(0.558072924613953, leftMatrix[leftMatrix.Count - 1], 5);
+                Assert.Equal(0.270811557769775, rightMatrix[0], 5);
+                Assert.Equal(0.376706808805466, rightMatrix[rightMatrix.Count - 1], 5);
             }
             // Read the test data set as an IDataView
             var testData = reader.Load(new MultiFileSource(GetDataPath(TestDatasets.trivialMatrixFactorization.testFilename)));
@@ -129,7 +129,7 @@ public void MatrixFactorizationSimpleTrainAndPredict()
             if (RuntimeInformation.IsOSPlatform(OSPlatform.Linux))
             {
                 // Linux case
-                var expectedUnixL2Error = 0.612974867782832; // Linux baseline
+                var expectedUnixL2Error = 0.610332110253861; // Linux baseline
                 Assert.InRange(metrices.MeanSquaredError, expectedUnixL2Error - linuxTolerance, expectedUnixL2Error + linuxTolerance);
             }
             else if (RuntimeInformation.IsOSPlatform(OSPlatform.OSX))
@@ -142,7 +142,7 @@ public void MatrixFactorizationSimpleTrainAndPredict()
             else if (RuntimeInformation.IsOSPlatform(OSPlatform.Windows))
             {
                 // Windows case
-                var expectedWindowsL2Error = 0.622283290742721; // Windows baseline
+                var expectedWindowsL2Error = 0.60226203382884; // Windows baseline
                 Assert.InRange(metrices.MeanSquaredError, expectedWindowsL2Error - windowsTolerance, expectedWindowsL2Error + windowsTolerance);
             }