diff --git a/lyra2/lyra2RE.cu b/lyra2/lyra2RE.cu
index 4cacbb9..e25aba4 100644
--- a/lyra2/lyra2RE.cu
+++ b/lyra2/lyra2RE.cu
@@ -6,8 +6,8 @@ extern "C" {
 #include "lyra2/Lyra2.h"
 }
 
-#include "miner.h"
-#include "cuda_helper.h"
+#include <miner.h>
+#include <cuda_helper.h>
 
 static uint64_t* d_hash[MAX_GPUS];
 static uint64_t* d_matrix[MAX_GPUS];
@@ -31,20 +31,6 @@ extern void groestl256_setTarget(const void *ptarget);
 extern uint32_t groestl256_cpu_hash_32(int thr_id, uint32_t threads, uint32_t startNounce, uint64_t *d_outputHash, int order);
 extern uint32_t groestl256_getSecNonce(int thr_id, int num);
 
-#ifdef _DEBUG
-#define TRACE(algo) { \
-	if (max_nonce == 1 && pdata[19] <= 1) { \
-		uint32_t* debugbuf = NULL; \
-		cudaMallocHost(&debugbuf, 8*sizeof(uint32_t)); \
-		cudaMemcpy(debugbuf, d_hash[thr_id], 8*sizeof(uint32_t), cudaMemcpyDeviceToHost); \
-		printf("lyra %s %08x %08x %08x %08x...\n", algo, swab32(debugbuf[0]), swab32(debugbuf[1]), \
-			swab32(debugbuf[2]), swab32(debugbuf[3])); \
-		cudaFreeHost(debugbuf); \
-	} \
-}
-#else
-#define TRACE(algo) {}
-#endif
 
 extern "C" void lyra2re_hash(void *state, const void *input)
 {
@@ -141,7 +127,6 @@ extern "C" int scanhash_lyra2(int thr_id, struct work* work, uint32_t max_nonce,
 		keccak256_cpu_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], order++);
 		lyra2_cpu_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], gtx750ti);
 		skein256_cpu_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], order++);
-		TRACE("S")
 
 		*hashes_done = pdata[19] - first_nonce + throughput;
 
diff --git a/lyra2/lyra2REv2.cu b/lyra2/lyra2REv2.cu
index c148372..15ca59d 100644
--- a/lyra2/lyra2REv2.cu
+++ b/lyra2/lyra2REv2.cu
@@ -7,9 +7,8 @@ extern "C" {
 #include "lyra2/Lyra2.h"
 }
 
-#include "miner.h"
-#include "cuda_helper.h"
-
+#include <miner.h>
+#include <cuda_helper.h>
 
 static uint64_t *d_hash[MAX_GPUS];
 static uint64_t* d_matrix[MAX_GPUS];
@@ -73,21 +72,6 @@ void lyra2v2_hash(void *state, const void *input)
 	memcpy(state, hashA, 32);
 }
 
-#ifdef _DEBUG
-#define TRACE(algo) { \
-	if (max_nonce == 1 && pdata[19] <= 1) { \
-		uint32_t* debugbuf = NULL; \
-		cudaMallocHost(&debugbuf, 32); \
-		cudaMemcpy(debugbuf, d_hash[thr_id], 32, cudaMemcpyDeviceToHost); \
-		printf("lyra2 %s %08x %08x %08x %08x...%08x... \n", algo, swab32(debugbuf[0]), swab32(debugbuf[1]), \
-			swab32(debugbuf[2]), swab32(debugbuf[3]), swab32(debugbuf[7])); \
-		cudaFreeHost(debugbuf); \
-	} \
-}
-#else
-#define TRACE(algo) {}
-#endif
-
 static bool init[MAX_GPUS] = { 0 };
 
 extern "C" int scanhash_lyra2v2(int thr_id, struct work* work, uint32_t max_nonce, unsigned long *hashes_done)
@@ -145,17 +129,11 @@ extern "C" int scanhash_lyra2v2(int thr_id, struct work* work, uint32_t max_nonc
 		int order = 0;
 
 		blake256_cpu_hash_80(thr_id, throughput, pdata[19], d_hash[thr_id], order++);
-		TRACE("blake  :");
 		keccak256_cpu_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], order++);
-		TRACE("keccak :");
 		cubehash256_cpu_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], order++);
-		TRACE("cube   :");
 		lyra2v2_cpu_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], order++);
-		TRACE("lyra2  :");
 		skein256_cpu_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], order++);
-		TRACE("skein  :");
 		cubehash256_cpu_hash_32(thr_id, throughput,pdata[19], d_hash[thr_id], order++);
-		TRACE("cube   :");
 
 		memset(work->nonces, 0, sizeof(work->nonces));
 		bmw256_cpu_hash_32(thr_id, throughput, pdata[19], d_hash[thr_id], work->nonces);