PhoenixMiner 6.2c: fastest Ethereum/Ethash miner with lowest devfee (Win/Linux)

[DrX]

Code:

[quote author=zlatarovv link=topic=2647654.msg55760704#msg55760704 date=1607198901]
[quote author=mraychev link=topic=2647654.msg55757515#msg55757515 date=1607173613]
[quote author=zlatarovv link=topic=2647654.msg55749355#msg55749355 date=1607082236]
Hello,

I switched to phoneix 5.3, but I notice an increased energy consultation.

I have platforms 2x6 shapshire rx 470 8g - 350 mh / s, 1900watts! at cleymore 15 - 1500watts
1x 5 rx 580 powercolor - 150 mh / s / 920watts - claymore 15 800watts

please advice and settings.

SAPPHIRE NITRO+ Radeon RX 470 8GB

2020.12.04: 03: 17: 30.750: main No CUDA driver found
2020.12.04: 03: 17: 33.584: main OpenCL driver version: 17.31.0
2020.12.04: 03: 17: 33.624: main Available GPUs for mining:
2020.12.04: 03: 17: 33.624: main GPU1: Radeon (TM) RX 470 Graphics (pcie 1), OpenCL 2.0, 8 GB VRAM, 32 CUs
2020.12.04: 03: 17: 33.624: main GPU2: Radeon (TM) RX 470 Graphics (pcie 2), OpenCL 2.0, 8 GB VRAM, 32 CUs
2020.12.04: 03: 17: 33.624: main GPU3: Radeon (TM) RX 470 Graphics (pcie 3), OpenCL 2.0, 8 GB VRAM, 32 CUs
2020.12.04: 03: 17: 33.624: main GPU4: Radeon (TM) RX 470 Graphics (pcie 4), OpenCL 2.0, 8 GB VRAM, 32 CUs
2020.12.04: 03: 17: 33.624: main GPU5: Radeon (TM) RX 470 Graphics (pcie 5), OpenCL 2.0, 8 GB VRAM, 32 CUs
2020.12.04: 03: 17: 33.624: main GPU6: Radeon (TM) RX 470 Graphics (pcie 6), OpenCL 2.0, 8 GB VRAM, 32 CUs
2020.12.04: 03: 17: 33.625: main ADL library initialized

RX 580 8gb powercolor
2020.12.04:13:18:02.423: main Phoenix Miner 5.3b Windows/msvc - Release build
2020.12.04:13:18:02.423: main No CUDA driver found
2020.12.04:13:18:03.769: main OpenCL driver version: 17.31.0
2020.12.04:13:18:03.790: main Available GPUs for mining:
2020.12.04:13:18:03.790: main GPU1: Radeon RX 580 Series (pcie 1), OpenCL 2.0, 8 GB VRAM, 36 CUs
2020.12.04:13:18:03.790: main GPU2: Radeon RX 580 Series (pcie 3), OpenCL 2.0, 8 GB VRAM, 36 CUs
2020.12.04:13:18:03.790: main GPU3: Radeon RX 580 Series (pcie 5), OpenCL 2.0, 8 GB VRAM, 36 CUs
2020.12.04:13:18:03.790: main GPU4: Radeon RX 580 Series (pcie 7), OpenCL 2.0, 8 GB VRAM, 36 CUs
2020.12.04:13:18:03.790: main GPU5: Radeon RX 580 Series (pcie 8), OpenCL 2.0, 8 GB VRAM, 36 CUs
2020.12.04:13:18:03.791: main ADL library initialized

Do I need to change the OpenCL driver or driver for card:

bat file:

@echo off
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_SINGLE_ALLOC_PERCENT 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_HEAP_SIZE 100
setx GPU_IMAGE_DMA 1
setx GPU_ENABLE_LARGE_ALLOCATION 1
setx GPU_FORCE_64BIT_PTR 0

PhoenixMiner.exe -pool eth-eu1.nanopool.org:9999 -wal
-amd
-acm
-mode 1
-hstats 1
-cclock 1150
-mclock 2040
-cvddc 900
-mvddc 850
-log 0
-tstop 85
-tstart 55
-tt 65
-fanmin 45
-fanmax 88
-cdm 0
-fcm 0
-gpus 1,2,3,4,5,6,7
-clKernel 1
-gser 5
-rvram -1
-wdog 1
-rmode 0
-powlim -50
-mt 2
I may have put something I shouldn't
They were not used for a very long time and worked only on Claymore. Please tell me which one to remove or put on to reduce electricity consumption

Settings of afterburner:
core voltage - /-98/
power limit - /-16/
temp. limit - /78/
cclock - 1150
mmclock - 2100/for rx580/; for rx470 - 2050




[/quote]

1. download and install and drivers 19.9
2. open and settings go to every gpu and witch from graphics to compute mode
this should work... I guess you are using the 17 blockchain beta driver now
[/quote]

Dear all,
I changed the drivers, but there is no change in power consumption?
If you have other suggestions, the cards consume over 110 watts
[/quote]

You set same settings in Afterburner and in PM? You don't need AB if you use PM to set the settings. What wattage does PM report?

[bobalazs]

Which wallet is the safest for Ethereum?
I tried geth but it didn't work for me.
How safe is the unofficial myetherwallet? or mycrypto?

I use myetherwallet and use coin import  function on guarda. It's very convenient that way.

[trinityfl]

wait update for 390 (-61)

[AzzAz]

Hello,

I switched to phoneix 5.3, but I notice an increased energy consultation.

I have platforms 2x6 shapshire rx 470 8g - 350 mh / s, 1900watts! at cleymore 15 - 1500watts
1x 5 rx 580 powercolor - 150 mh / s / 920watts - claymore 15 800watts

please advice and settings.

SAPPHIRE NITRO+ Radeon RX 470 8GB

2020.12.04: 03: 17: 30.750: main No CUDA driver found
2020.12.04: 03: 17: 33.584: main OpenCL driver version: 17.31.0
2020.12.04: 03: 17: 33.624: main Available GPUs for mining:
2020.12.04: 03: 17: 33.624: main GPU1: Radeon (TM) RX 470 Graphics (pcie 1), OpenCL 2.0, 8 GB VRAM, 32 CUs
2020.12.04: 03: 17: 33.624: main GPU2: Radeon (TM) RX 470 Graphics (pcie 2), OpenCL 2.0, 8 GB VRAM, 32 CUs
2020.12.04: 03: 17: 33.624: main GPU3: Radeon (TM) RX 470 Graphics (pcie 3), OpenCL 2.0, 8 GB VRAM, 32 CUs
2020.12.04: 03: 17: 33.624: main GPU4: Radeon (TM) RX 470 Graphics (pcie 4), OpenCL 2.0, 8 GB VRAM, 32 CUs
2020.12.04: 03: 17: 33.624: main GPU5: Radeon (TM) RX 470 Graphics (pcie 5), OpenCL 2.0, 8 GB VRAM, 32 CUs
2020.12.04: 03: 17: 33.624: main GPU6: Radeon (TM) RX 470 Graphics (pcie 6), OpenCL 2.0, 8 GB VRAM, 32 CUs
2020.12.04: 03: 17: 33.625: main ADL library initialized

RX 580 8gb powercolor
2020.12.04:13:18:02.423: main Phoenix Miner 5.3b Windows/msvc - Release build
2020.12.04:13:18:02.423: main No CUDA driver found
2020.12.04:13:18:03.769: main OpenCL driver version: 17.31.0
2020.12.04:13:18:03.790: main Available GPUs for mining:
2020.12.04:13:18:03.790: main GPU1: Radeon RX 580 Series (pcie 1), OpenCL 2.0, 8 GB VRAM, 36 CUs
2020.12.04:13:18:03.790: main GPU2: Radeon RX 580 Series (pcie 3), OpenCL 2.0, 8 GB VRAM, 36 CUs
2020.12.04:13:18:03.790: main GPU3: Radeon RX 580 Series (pcie 5), OpenCL 2.0, 8 GB VRAM, 36 CUs
2020.12.04:13:18:03.790: main GPU4: Radeon RX 580 Series (pcie 7), OpenCL 2.0, 8 GB VRAM, 36 CUs
2020.12.04:13:18:03.790: main GPU5: Radeon RX 580 Series (pcie , OpenCL 2.0, 8 GB VRAM, 36 CUs
2020.12.04:13:18:03.791: main ADL library initialized

Do I need to change the OpenCL driver or driver for card:

bat file:

@echo off
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_SINGLE_ALLOC_PERCENT 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_HEAP_SIZE 100
setx GPU_IMAGE_DMA 1
setx GPU_ENABLE_LARGE_ALLOCATION 1
setx GPU_FORCE_64BIT_PTR 0

PhoenixMiner.exe -pool eth-eu1.nanopool.org:9999 -wal
-amd
-acm
-mode 1
-hstats 1
-cclock 1150
-mclock 2040
-cvddc 900
-mvddc 850
-log 0
-tstop 85
-tstart 55
-tt 65
-fanmin 45
-fanmax 88
-cdm 0
-fcm 0
-gpus 1,2,3,4,5,6,7
-clKernel 1
-gser 5
-rvram -1
-wdog 1
-rmode 0
-powlim -50
-mt 2
I may have put something I shouldn't
They were not used for a very long time and worked only on Claymore. Please tell me which one to remove or put on to reduce electricity consumption

Settings of afterburner:
core voltage - /-98/
power limit - /-16/
temp. limit - /78/
cclock - 1150
mmclock - 2100/for rx580/; for rx470 - 2050

1. download and install and drivers 19.9
2. open and settings go to every gpu and witch from graphics to compute mode
this should work... I guess you are using the 17 blockchain beta driver now

Dear all,
I changed the drivers, but there is no change in power consumption?
If you have other suggestions, the cards consume over 110 watts

cclock 1131 or 1141 or 1151 ( check watts ), mclock 2100, cvddc 860, mvddc 875, fanmin and fanmax remove, afterburner remove,, use tt 63 and ttli 70, and 20.4.2 driver ( that depends of windows version ) ...  - mine 470's do over 30Mh and under 110w ... dont know about 580 Powercolor's

[dastardlyman12]

lost radeon VII hashrate when we hit dag 379
amd driver 20.3.1
something peculier has just happened.
this may help vega and radeon vii owners
i updated my radeonvii rig to latest windows 10 called 20H2
then went back into updates and did the 20h2 optional update thingy.
it takes a few minutes and two reboots required
started miner - got 80 mhash per card - the expected reduced rate.
waited three hours and -
just now the rig stopped itself.
i restarted it and phoenix saying "unknown opencl version" but ive got my missing hashes back.
windows has updated something to do with opencl and the hashes are back. 83mhash per card
hmm
cant say i undertstand this at all.
 am getting shares - seems to be ok - no problems

i didnt update opencl - windows did itself.

dag 380 due in two hours wonder what will happen then?

[cryptocryptic]

With Adrenaline 20.4.2, I am only getting 10.5 to 11 MH/s in my Rx580 8GB cards in PhoenixMiner. Tried the usual steps, used DDU to reinstall drivers, did a fresh Windows install as well. No improvements, the hash rate stays at around 11 only.  Was getting 30 MH/s with the 2017 beta drivers all this while.

The only reason I upgraded the drivers was Claymore died and when I tried mining with PhoenixMiner, while the hashrate came upto 30 MH/s, it wouldn't undervolt using cvvdc and was causing too much power to be drawn.

What am I missing?

[UnclWish]

With Adrenaline 20.4.2, I am only getting 10.5 to 11 MH/s in my Rx580 8GB cards in PhoenixMiner. Tried the usual steps, used DDU to reinstall drivers, did a fresh Windows install as well. No improvements, the hash rate stays at around 11 only.  Was getting 30 MH/s with the 2017 beta drivers all this while.

The only reason I upgraded the drivers was Claymore died and when I tried mining with PhoenixMiner, while the hashrate came upto 30 MH/s, it wouldn't undervolt using cvvdc and was causing too much power to be drawn.

What am I missing?

You forgot to change GPU to compute mode.

[bluspirit]

With Adrenaline 20.4.2, I am only getting 10.5 to 11 MH/s in my Rx580 8GB cards in PhoenixMiner. Tried the usual steps, used DDU to reinstall drivers, did a fresh Windows install as well. No improvements, the hash rate stays at around 11 only.  Was getting 30 MH/s with the 2017 beta drivers all this while.

The only reason I upgraded the drivers was Claymore died and when I tried mining with PhoenixMiner, while the hashrate came upto 30 MH/s, it wouldn't undervolt using cvvdc and was causing too much power to be drawn.

What am I missing?

In AMD global settings try to change GPU Workload from Graphics to Compute for each card.
Use the -rxboost 1 option in PM command and run the miner as administrator.

[ivomm]

wait update for 390 (-61)

For now, you can try nanominer or gminer. Some cards are having higher number of stale shares with nanominer like 15%, while with gminer they are around 2.5%. For some reason gminer is not working on smos (althoкgh i didn't test the newest version of smos), but on hiveos it is working well. Not so well as Claymore, but good enough.

[Malefol7]

hi, I'm new to the forum, I run a pool with core-geth 1.11.19, and I run 5.3b but I get incorrect shares, why?
amd 20.4.x
https://etc.cominers.org/screen.png

[cryptocryptic]

With Adrenaline 20.4.2, I am only getting 10.5 to 11 MH/s in my Rx580 8GB cards in PhoenixMiner. Tried the usual steps, used DDU to reinstall drivers, did a fresh Windows install as well. No improvements, the hash rate stays at around 11 only.  Was getting 30 MH/s with the 2017 beta drivers all this while.

The only reason I upgraded the drivers was Claymore died and when I tried mining with PhoenixMiner, while the hashrate came upto 30 MH/s, it wouldn't undervolt using cvvdc and was causing too much power to be drawn.

What am I missing?

In AMD global settings try to change GPU Workload from Graphics to Compute for each card.
Use the -rxboost 1 option in PM command and run the miner as administrator.

This worked! Thank you so much.

[cscheat]

I would really love to see a screen shot of ANYONE that can pump out more than 30 MH/card on (12+ per rig) RX 580 cards and hash stable for more than 24 hrs. No matter what I try it just will not happen. As soon as I push to 30 or higher, the rig will start to crash. I used to tune individual cards and get higher hash rate but that function has been broken since forever ago. So I would really love to see if anyone out there is actually able to achieve this.

Here You Go man: My 12 Card RIG with 8* RX580 8GB, RX5700, 3* RX5700XT
https://i.imgur.com/dwdqyFz.jpeg

I appreciate your screen shot but it only shows your rig at 47 min. I was asking if anyone can show a rig running 30+ MH for more than 24 hr.


Side note: Did PM ever fix the problem with not being able to set cclock and mclock for individual cards?

one of my rig...

https://i.imgur.com/dkQk1Gm.jpg

[zeef]

I would really love to see a screen shot of ANYONE that can pump out more than 30 MH/card on (12+ per rig) RX 580 cards and hash stable for more than 24 hrs. No matter what I try it just will not happen. As soon as I push to 30 or higher, the rig will start to crash. I used to tune individual cards and get higher hash rate but that function has been broken since forever ago. So I would really love to see if anyone out there is actually able to achieve this.

Here You Go man: My 12 Card RIG with 8* RX580 8GB, RX5700, 3* RX5700XT
https://i.imgur.com/dwdqyFz.jpeg

I appreciate your screen shot but it only shows your rig at 47 min. I was asking if anyone can show a rig running 30+ MH for more than 24 hr.


Side note: Did PM ever fix the problem with not being able to set cclock and mclock for individual cards?

one of my rig...

https://i.imgur.com/dkQk1Gm.jpg

What OC settings you have to get so low power consumption?

[vydelek]

Good day,
for multiple RIGs at the same setting everything is OK but for three rigs it happened that PM stopped mining.

Please what is the problem, what command to try for DAG generation.

PM VERZE 5.2e


2020.12.06:16:26:50.440: main GPUs: 1: 31.047 MH/s (71) 2: 31.580 MH/s (94) 3: 31.585 MH/s (65) 4: 31.299 MH/s (71) 5: 30.667 MH/s (77) 6: 31.581 MH/s (79) 7: 31.448 MH/s (78)
2020.12.06:16:26:51.705: eths Eth: Received: {"id":0,"jsonrpc":"2.0","result":["0x235323ee2760cc24f0f13eda8a0d707624c68dee7fbb17baaf3582abf9da6f32","0x24a739dd7e88652c7b9c8b21ab1ea88a6a1bc824232e6ffb030387c1518290d7","0x0000000112e0be826d694b2e62d01511f12a6061fbaec8bc02357593e70e52ba","0xadf33f"]}
2020.12.06:16:26:51.706: eths Eth: New job #235323ee from ssl://eu1.ethermine.org:5555; diff: 4000MH
2020.12.06:16:26:52.658: eths Eth: Received: {"id":0,"jsonrpc":"2.0","result":["0x55ef1a72e2937b6f4c7167c95f610335ae78e35b1f1200763b8a8cacda35b9c5","0x0f019d165efddfcfc233aab381f260fcc26c8bcf497dcffc4b78fd0c5223b3d1","0x0000000112e0be826d694b2e62d01511f12a6061fbaec8bc02357593e70e52ba","0xadf340"]}
2020.12.06:16:26:52.658: eths Eth: New job #55ef1a72 from ssl://eu1.ethermine.org:5555; diff: 4000MH
2020.12.06:16:26:52.941: eths Eth: Received: {"id":0,"jsonrpc":"2.0","result":["0xad96f4a8851530a0bf859edb3ee2d161d66b8063d16e56068823a36b757093eb","0x0f019d165efddfcfc233aab381f260fcc26c8bcf497dcffc4b78fd0c5223b3d1","0x0000000112e0be826d694b2e62d01511f12a6061fbaec8bc02357593e70e52ba","0xadf340"]}
2020.12.06:16:26:52.942: eths Eth: New job #ad96f4a8 from ssl://eu1.ethermine.org:5555; diff: 4000MH
2020.12.06:16:26:52.960: GPU7 GPU7: Starting up... (0)
2020.12.06:16:26:52.961: GPU7 GPU7: Generating ethash light cache for epoch #380
2020.12.06:16:26:52.963: GPU3 GPU3: Starting up... (0)
2020.12.06:16:26:52.983: GPU4 GPU4: Starting up... (0)
2020.12.06:16:26:53.101: GPU2 GPU2: Starting up... (0)
2020.12.06:16:26:53.103: GPU5 GPU5: Starting up... (0)
2020.12.06:16:26:53.140: GPU1 GPU1: Starting up... (0)
2020.12.06:16:26:53.160: GPU6 GPU6: Starting up... (0)
2020.12.06:16:26:55.457: main Eth speed: 0.000 MH/s, shares: 535/0/0, time: 2:36
2020.12.06:16:26:55.457: main GPUs: 1: 0.000 MH/s (71) 2: 0.000 MH/s (94) 3: 0.000 MH/s (65) 4: 0.000 MH/s (71) 5: 0.000 MH/s (77) 6: 0.000 MH/s (79) 7: 0.000 MH/s (78)
2020.12.06:16:26:56.748: eths Eth: Received: {"id":0,"jsonrpc":"2.0","result":["0xa4159c8a17bd9674060a4282c2180933adb295c8fe38ce4683e56ec3a3d62f0e","0x0f019d165efddfcfc233aab381f260fcc26c8bcf497dcffc4b78fd0c5223b3d1","0x0000000112e0be826d694b2e62d01511f12a6061fbaec8bc02357593e70e52ba","0xadf340"]}
2020.12.06:16:26:56.748: eths Eth: New job #a4159c8a from ssl://eu1.ethermine.org:5555; diff: 4000MH
2020.12.06:16:26:57.131: GPU7 Light cache generated in 4.2 s (15.2 MB/s)
2020.12.06:16:26:57.463: eths Eth: Send: {"id":5,"jsonrpc":"2.0","method":"eth_getWork","params":[]}

2020.12.06:16:26:57.477: eths Eth: Received: {"id":6,"jsonrpc":"2.0","result":true}
2020.12.06:16:26:57.477: eths Eth: Received: {"id":0,"jsonrpc":"2.0","result":["0xa4159c8a17bd9674060a4282c2180933adb295c8fe38ce4683e56ec3a3d62f0e","0x0f019d165efddfcfc233aab381f260fcc26c8bcf497dcffc4b78fd0c5223b3d1","0x0000000112e0be826d694b2e62d01511f12a6061fbaec8bc02357593e70e52ba","0xadf340"]}
2020.12.06:16:27:00.479: main Eth speed: 0.000 MH/s, shares: 535/0/0, time: 2:36
2020.12.06:16:27:00.479: main GPUs: 1: 0.000 MH/s (71) 2: 0.000 MH/s (94) 3: 0.000 MH/s (65) 4: 0.000 MH/s (71) 5: 0.000 MH/s (77) 6: 0.000 MH/s (79) 7: 0.000 MH/s (78)
2020.12.06:16:27:00.870: eths Eth: Received: {"id":0,"jsonrpc":"2.0","result":["0xf3ae18a4ef2381e6fa0c8a14b4f4525d6151597834959fc4d0e71abb0cb4a408","0x0f019d165efddfcfc233aab381f260fcc26c8bcf497dcffc4b78fd0c5223b3d1","0x0000000112e0be826d694b2e62d01511f12a6061fbaec8bc02357593e70e52ba","0xadf340"]}
2020.12.06:16:27:00.870: eths Eth: New job #f3ae18a4 from ssl://eu1.ethermine.org:5555; diff: 4000MH
2020.12.06:16:27:02.132: GPU2 GPU2: Generating DAG for epoch #380
2020.12.06:16:27:02.132: GPU4 GPU4: Generating DAG for epoch #380
2020.12.06:16:27:02.132: GPU3 GPU3: Generating DAG for epoch #380
2020.12.06:16:27:02.133: GPU1 GPU1: Generating DAG for epoch #380
2020.12.06:16:27:02.133: GPU6 GPU6: Generating DAG for epoch #380
2020.12.06:16:27:02.133: GPU5 GPU5: Generating DAG for epoch #380
2020.12.06:16:27:02.133: GPU7 GPU7: Generating DAG for epoch #380
2020.12.06:16:27:02.887: main GPU1: 54C 33% 42W, GPU2: 52C 32% 32W, GPU3: 51C 30% 42W, GPU4: 52C 32% 40W, GPU5: 52C 31% 41W, GPU6: 50C 30% 40W, GPU7: 47C 28% 33W
GPUs power: 270.1 W
2020.12.06:16:27:05.491: main Eth speed: 0.000 MH/s, shares: 535/0/0, time: 2:36
2020.12.06:16:27:05.491: main GPUs: 1: 0.000 MH/s (71) 2: 0.000 MH/s (94) 3: 0.000 MH/s (65) 4: 0.000 MH/s (71) 5: 0.000 MH/s (77) 6: 0.000 MH/s (79) 7: 0.000 MH/s (78)
2020.12.06:16:27:07.463: eths Eth: Send: {"id":5,"jsonrpc":"2.0","method":"eth_getWork","params":[]}

2020.12.06:16:27:07.463: eths Eth: Send: {"id":6,"jsonrpc":"2.0","method":"eth_submitHashrate","params":["0xc633a5c","0x736bfa2f7e89ddc019c7a171628695c5f085dea0c0d1c0313f0aa27e4b3fc10a"]}

THE END - NO MINING - ERROR - PLEASE HELP

[waywoos]

guys the 4gb days are done with eth.
mine etc or beam or ubiq or other coins that require a less than 4gb dag
there is no solution besides going to linux or other non windows OS for a few more weeks at best

[vydelek]

Please why PM does not auto reset when GK crashes on -rmode 2 command when running as administrator? WIN10

[casslo]

guys the 4gb days are done with eth.
mine etc or beam or ubiq or other coins that require a less than 4gb dag
there is no solution besides going to linux or other non windows OS for a few more weeks at best

just move to a usb OS like hiveOS or other, its pretty simple tbh, i setup it very quick without any problem last week and 4gb are working there 100%

[waywoos]

guys the 4gb days are done with eth.
mine etc or beam or ubiq or other coins that require a less than 4gb dag
there is no solution besides going to linux or other non windows OS for a few more weeks at best

just move to a usb OS like hiveOS or other, its pretty simple tbh, i setup it very quick without any problem last week and 4gb are working there 100%

Yeah it’s temporary

[Vladegg]

Please why PM does not auto reset when GK crashes on -rmode 2 command when running as administrator? WIN10

Win 10 LTSC 17763, Win 10 Pro 19041 - Phoenixminer 5.3b stuck on errors and system can't restart. Doesn't matter PM under Administrator or not.

[goxed]

Hey PhoenixMiner, ever since the last ETH Epoch and the new 3.98GB DAG, I am seeing a slowdown on all my mining graphics cards. It varies from 5 - 10% depending on the card type and other settings. I am using version 5.3b. Any explanations as what could be the reason for this slowdown?

The slowdown remain after re-booting.