diff --git a/CHANGELOG.md b/CHANGELOG.md
index 543db9ac..ac3ba707 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -25,10 +25,11 @@
 ### Bug Fixes
 - extractor: fix vivisect loop detection corner case #1310 @mr-tz
 - match: extend OS characteristic to match OS_ANY to all supported OSes #1324 @mike-hunhoff
-- explorer: fix exception when plugin loaded in IDA hosted under idat #1341 @mike-hunhoff
 - extractor: fix IDA and vivisect string and bytes features overlap and tests #1327 #1336 @xusheng6 
 
 ### capa explorer IDA Pro plugin
+- fix exception when plugin loaded in IDA hosted under idat #1341 @mike-hunhoff
+- improve embedded PE detection performance and reduce FP potential #1344 @mike-hunhoff
 
 ### Development
 
diff --git a/capa/features/extractors/ida/file.py b/capa/features/extractors/ida/file.py
index 0d380a88..532d5a89 100644
--- a/capa/features/extractors/ida/file.py
+++ b/capa/features/extractors/ida/file.py
@@ -21,12 +21,14 @@ from capa.features.file import Export, Import, Section, FunctionName
 from capa.features.common import FORMAT_PE, FORMAT_ELF, Format, String, Feature, Characteristic
 from capa.features.address import NO_ADDRESS, Address, FileOffsetAddress, AbsoluteVirtualAddress
 
+MAX_OFFSET_PE_AFTER_MZ = 0x200
+
 
 def check_segment_for_pe(seg: idaapi.segment_t) -> Iterator[Tuple[int, int]]:
     """check segment for embedded PE
 
     adapted for IDA from:
-    https://github.com/vivisect/vivisect/blob/7be4037b1cecc4551b397f840405a1fc606f9b53/PE/carve.py#L19
+    https://github.com/vivisect/vivisect/blob/91e8419a861f49779f18316f155311967e696836/PE/carve.py#L25
     """
     seg_max = seg.end_ea
     mz_xor = [
@@ -40,13 +42,14 @@ def check_segment_for_pe(seg: idaapi.segment_t) -> Iterator[Tuple[int, int]]:
 
     todo = []
     for mzx, pex, i in mz_xor:
+        # find all segment offsets containing XOR'd "MZ" bytes
         for off in capa.features.extractors.ida.helpers.find_byte_sequence(seg.start_ea, seg.end_ea, mzx):
             todo.append((off, mzx, pex, i))
 
     while len(todo):
         off, mzx, pex, i = todo.pop()
 
-        # The MZ header has one field we will check e_lfanew is at 0x3c
+        # MZ header has one field we will check e_lfanew is at 0x3c
         e_lfanew = off + 0x3C
 
         if seg_max < (e_lfanew + 4):
@@ -54,6 +57,10 @@ def check_segment_for_pe(seg: idaapi.segment_t) -> Iterator[Tuple[int, int]]:
 
         newoff = struct.unpack("<I", capa.features.extractors.helpers.xor_static(idc.get_bytes(e_lfanew, 4), i))[0]
 
+        # assume XOR'd "PE" bytes exist within threshold
+        if newoff > MAX_OFFSET_PE_AFTER_MZ:
+            continue
+
         peoff = off + newoff
         if seg_max < (peoff + 2):
             continue
@@ -61,9 +68,6 @@ def check_segment_for_pe(seg: idaapi.segment_t) -> Iterator[Tuple[int, int]]:
         if idc.get_bytes(peoff, 2) == pex:
             yield off, i
 
-        for nextres in capa.features.extractors.ida.helpers.find_byte_sequence(off + 1, seg.end_ea, mzx):
-            todo.append((nextres, mzx, pex, i))
-
 
 def extract_file_embedded_pe() -> Iterator[Tuple[Feature, Address]]:
     """extract embedded PE features