Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux)

[bategojko74]

I know you can take a log file and if you have a long enough mining duration say 1 month, then you can easily add all the accepted+rejected+stale and get an accurate measurement and being within +- 1% is fairly close.

I think the 1% loss you are talking about is not related to the Claymore software but most likely due to the GPU drivers. I know with my AMD RX series GPUs there is a time when some GPUs hash a little slower for a few seconds and then resume the optimal speed. Been trying to figure out why it happens but couldn't and just left it the way it was because the speed loss was minimal. So find in your logs if your speed was always constant or did it slow down from time to time. You can probably find this easily with MS Excel when you load your log file.

Nobody is talking about speed fluctuations here. The averaged stated speed incorporates them. If miner does not work at some moment it prints 0.000 Mh/s in console (and log) and this zero is heaped when calculating the averaged speed and the number of samples is increased by one for this zero. Come on man use a little the last floor of your body. I talk that this averaged stated speed in console and in the log is 1.5% higher than the real one which can be calculated from the found shares (locally, not at the pools side), difficulty and the time that the miner has worked.

I'm trying to follow but it's a bit confusing. So now you are claiming that the speed that the miner displays in console and reports to the pool is 1.5% higher than the actual one? and is that somehow related to your previous claim of Claymore taking higher than advertised devfee?

I was wrong that Claymore is taking higher than the advertised devfee. But I am 1000% sure that Claymore shows and reports to pools (if he reports what he shows in console) about 1.5% higher hash rate than the real one.

Ohh I see.
If you are into this kind of action can you maybe do all of us a favor and perform the same analysis on this one https://bitcointalk.org/index.php?topic=2647654.0
And maybe post here with numbers and such?
I know for myself I'd be very much interested to see the actual numbers. Thx.

Partial results for PM 4.6C after two days mining with 10 RX570 + 1 RX580 and 14889 shares found (14799 shares + 90 devfee shares):
https://drive.google.com/open?id=1luXj5zlyeFdXeQ4QL9GcS1t7bSUPzAQd

Compile AverageSpeed.cpp with visual studio

Or if you use GCC compiler:
comment row 4:
//#include "stdafx.h"

and compile it with:
g++ -o AverageSpeed.exe AverageSpeed.cpp

and run it with the following parameters:
AverageSpeed.exe 3 main Eth speed: log20190929_174105.txt

It will produce the averaged stated speed: 316.689 MH/s
The real effective speed is: 14889 (total shares) * 4,000,000,000 (ethermine difficulty) / 188886 (the seconds that the miner has worked) = 315.301 MH/s
So PM shows (316.689 / 315,301 - 1) * 100 = 0.44% higher hashrate than the real effective one. In comparision Claymore shows about 1.5% higher than the real one.
Will update the results when one week ends but 14889 found shares seems to me statistically representative test that will probably produce error under 0.1%. So I expect this 0.44% to finish between 0.34%-0.54% after 5 days.

this is interesting.

Final results for how much higher stated hashrate Phoenix miner shows than the real effective hashrate:
https://drive.google.com/open?id=1UwrGhi0y6eLW-Utw3U9v0pjKCdj0zDam

Total shares: 35923 (35703 + 220 devfee shares)

The averaged stated speed is: 316.7 MH/s
The real effective speed is: 35923 (total shares) * 4,000,000,000 (ethermine difficulty) / 455,640 (the seconds that the miner has worked) = 315.363 MH/s
So PM shows (316.7 / 315,363 - 1) * 100 = 0.42% higher hashrate than the real effective one. For comparision Claymore shows about 1.5% higher than the real one, which is 3 times higher than PM.

[Bittap]

Really doesn't matter. You just need to have at least 1 card that is large enough and able to load the dag alone. I have 6 rigs this is functioning on, all were 6x Rx470 4GB rigs, but then I took out a card in each and replaced each slot with an Rx580 8GB once I discovered the fix. Distributed the set ups so every 4GB rig has at least one 8GB card to load the dag. I have not confirmed if an Nvidia card can assist another Nvidia card in this way, I don't any under 8GB however the even stranger thing about all of this is that a p104-100 8GB will assist my Rx470 4GBs and they will mine.

You will need at least 1 GPU to load the DAG plus the Windows data system files on the GDDR. But you will still need to have at least 4GB for the other GPUs. What I mean is that you can't have 1 GPU that has 8GB GDDR and have the rest being 3GB GDDR GPUs. It won't work I tried.

The way the Claymore software works is that it needs to load the entire DAG onto each and every GPU. It doesn't just load it only on the first GPU. You can look at the resource history monitor and you will see that the memory for each GPU is completely used up. So this will still only work with any GPUs greater than 4GB but anything 3GB or less is obsolete in mining ETH right now.

What cards are you testing? This may not work for Nvidia 3Gbs as I stated I only got it to work with a Nvidia p104 assisting a Rx470 4Gba.

The way I discovered this fix was half my farm, about 6 rigs with all 4Gb cards, went down and stopped generating the dag on both eth and etc. I realized that one of my mixed rigs with a 8gb worked fine. I tested it by removing the 8Gb. All of sudden it didn't mine. I tested this on the other rigs by dismantling a 580 8Gb rig and distributing the cards to each of the 470 rigs. This works flawlessly.

As for my r7 200 series I have a desktop build running in Windows 10 legacy mode. That build I can install my 580 8Gb or other Polaris cards with enough memory for dag, and sit the R7 200 next to it. The dag loads perfectly fine and mines. Takes about 57 seconds to load the dag and mines at a pitiful 1.2Mhs.

I noticed this a few years back as I mentioned but now with the dag approaching some cards thresholds again and clearly mine, I rehashed this. (pun intended)

Legacy and UEFI mode have alot to do with how pcie slots will interact with you cpu and motherboards chipset. UEFI is going to give you the latest pcie options and most lanes. Legacy is a much older way of installing Windows and restricts pcie lanes available and other features included in the bios and operating system.

[RAKOT]

Hello!

How could i run two instances of CM 15 simultaneously ? I have miners in different dirs, using CM driver for straps feature , but second miner tells 'driver in use' and runs at not full speed.

[Viaggiatore]

@Claymore, I don't know if anyone already had this problem. I have a 8 x RX570 8GB rig mining solo ETH, randomly one card (any card) drops to 22Mhs for minutes and then it starts again going to full speed. I have been seeing this for long time now, any advice?

Thanks

[MATHReX]

Hello!

How could i run two instances of CM 15 simultaneously ? I have miners in different dirs, using CM driver for straps feature , but second miner tells 'driver in use' and runs at not full speed.

You can run two instances of the miner but it will impact the hashrate.
Use proper parameters in both the bat files to include the GPUs that you want to run in a particular instance.
Moreover, the straps feature requires it to run with the Administrator and it modifies the straps of the device on the fly but it won't be able to do in the second instance.
You can try running the second instance too with Admin but I am not sure if it will work.

[MATHReX]

@Claymore, I don't know if anyone already had this problem. I have a 8 x RX570 8GB rig mining solo ETH, randomly one card (any card) drops to 22Mhs for minutes and then it starts again going to full speed. I have been seeing this for long time now, any advice?

Thanks

Does the fluctuation of the hash rate on the one card reflect in the pool?
I read it somewhere that when a correct hash is calculated by the VGA and sent to the pool, it may affect the hash rate of the card for just a moment.

Also, try to remove any OC that you have on that card and see if it solves the issue.

[amishmanish]

--snip--

Right now I can't test this out since all my cards are still working. But if you are getting issues, try and buy 1 GPU which is 8GB like the RX 470 8GB which you can probably buy for like $75 these days and try this out and it should work. Just make sure to put it in the 0 PCIe slot so that GPU loads the most of the DAG plus the Windows data resources and the rest should fit on the other 4GB GPUs.

Either this or just try Linux. There are distros available these days which are plug n play pretty much.

Could you please share where are you quoting these prices for an 8GB GPU from? Are those brand-new with warranty or used ones?? It'd be really helpful if you can tell about your experience of using these low priced hardware (if its used).
Anybody else reading this who knows about the supplies, do share.

[dougBTC]

Do we see AMD releasing new 5700 XT drivers to allow us to use all 40 compute units?   Miner is only seeing 20.

Thank you for the great software!

[MATHReX]

Do we see AMD releasing new 5700 XT drivers to allow us to use all 40 compute units?   Miner is only seeing 20.

Thank you for the great software!

It might just be a syntax error.
In actual, It is using all the computational power there is to get the max possible hash rate.
I have seen people getting 62 MH/s with 5700 XT.

[Bathmat]

Do we see AMD releasing new 5700 XT drivers to allow us to use all 40 compute units?   Miner is only seeing 20.

Thank you for the great software!

It might just be a syntax error.
In actual, It is using all the computational power there is to get the max possible hash rate.
I have seen people getting 62 MH/s with 5700 XT.

Everything I've seen boasting greater than 54Mh/s have been changing the OpenCL worksize and causing invalid shares that reduce effective hashrate back down to around 50-54Mh/s.

[RAKOT]

Hello!

How could i run two instances of CM 15 simultaneously ? I have miners in different dirs, using CM driver for straps feature , but second miner tells 'driver in use' and runs at not full speed.

You can run two instances of the miner but it will impact the hashrate.
Use proper parameters in both the bat files to include the GPUs that you want to run in a particular instance.
Moreover, the straps feature requires it to run with the Administrator and it modifies the straps of the device on the fly but it won't be able to do in the second instance.
You can try running the second instance too with Admin but I am not sure if it will work.

Well , i think its not good. New features are fantastic but itwork ( i meen 2 CM on same rig) before, why not now ? -> @Claymore.

[fredeq]

Everything I've seen boasting greater than 54Mh/s have been changing the OpenCL worksize and causing invalid shares that reduce effective hashrate back down to around 50-54Mh/s.

Exactly that. 1300/915@0.85 gives solid 51.5mh/s. If you increase worksize then reject rate increases to 10-20%

@claymore
Strange stuff regarding 380 hashrate reporting with v15. Just updated drivers from 18.6.2 to 19.10.1 and miner reports 380 hashrate at 37mh/s+.
Phoenix is still fine with ~19mh/s, which is the true hashrate of the device.

Fury hashrate reporting is fine. Not sure if it happened with earlier versions of the miner or previous 19.x drivers.

[bategojko74]

Everything I've seen boasting greater than 54Mh/s have been changing the OpenCL worksize and causing invalid shares that reduce effective hashrate back down to around 50-54Mh/s.

Exactly that. 1300/915@0.85 gives solid 51.5mh/s. If you increase worksize then reject rate increases to 10-20%

@claymore
Strange stuff regarding 380 hashrate reporting with v15. Just updated drivers from 18.6.2 to 19.10.1 and miner reports 380 hashrate at 37mh/s+.
Phoenix is still fine with ~19mh/s, which is the true hashrate of the device.

Fury hashrate reporting is fine. Not sure if it happened with earlier versions of the miner or previous 19.x drivers.

Decrease 0.85 to 0.75. You don't need 0.85

[fredeq]

Decrease 0.85 to 0.75. You don't need 0.85

Last time I checked 0.80 was not stable, but will recheck. Unlucky at chip lottery?

[nick_fc]

guys, last time was useng v12.0
stopped mining because 3gb GPUs stopped mining.
Now I see latest version is v15.
It says "no dag files"
is that mean I can use again my 3gb cards for ETH mining? Or 3gb are still unusable for ETH/ETC?

[adaseb]

--snip--

Right now I can't test this out since all my cards are still working. But if you are getting issues, try and buy 1 GPU which is 8GB like the RX 470 8GB which you can probably buy for like $75 these days and try this out and it should work. Just make sure to put it in the 0 PCIe slot so that GPU loads the most of the DAG plus the Windows data resources and the rest should fit on the other 4GB GPUs.

Either this or just try Linux. There are distros available these days which are plug n play pretty much.

Could you please share where are you quoting these prices for an 8GB GPU from? Are those brand-new with warranty or used ones?? It'd be really helpful if you can tell about your experience of using these low priced hardware (if its used).
Anybody else reading this who knows about the supplies, do share.

The prices are obviously for used units. And in my area it's all over Craiglists and eBay. The 8GB variants go for maybe $10 over the 4GB variants. Basically even the people with 8GB GPUs don't want to mine any more since the profitability is so low.

Currently I am just waiting to see what will happen to ProgPOW, only reason why I didn't eBay my GPUs just yet. If we can get a decent price increase followed by an ProgPOW change in the new year then mining might be worth while again but right now I am getting like a dime a day and just keeping them turned off.

[Viaggiatore]

@Claymore, I don't know if anyone already had this problem. I have a 8 x RX570 8GB rig mining solo ETH, randomly one card (any card) drops to 22Mhs for minutes and then it starts again going to full speed. I have been seeing this for long time now, any advice?

Thanks

Does the fluctuation of the hash rate on the one card reflect in the pool?
I read it somewhere that when a correct hash is calculated by the VGA and sent to the pool, it may affect the hash rate of the card for just a moment.

Also, try to remove any OC that you have on that card and see if it solves the issue.

Yes, it does reflect the card pool. With Phoenix and same OC settings I don't have any of this issue. Card are running really cold, so it is also out a question that cards are slowing down because of heat. It is not just a moment, I'm talking about minutes slowing down.

[MATHReX]

guys, last time was useng v12.0
stopped mining because 3gb GPUs stopped mining.
Now I see latest version is v15.
It says "no dag files"
is that mean I can use again my 3gb cards for ETH mining? Or 3gb are still unusable for ETH/ETC?

I am not a hundred percent sure what "No Dag Files" means but you will never be able to mine ETH/ETC using 3GB cards as DAG for both of it is above 3GB.
https://investoon.com/tools/dag_size

Yes, it does reflect the card pool. With Phoenix and same OC settings I don't have any of this issue. Card are running really cold, so it is also out a question that cards are slowing down because of heat. It is not just a moment, I'm talking about minutes slowing down.

Can you paste the complete claymore bat file text here?

[user0071]

I'm having problems getting Claymore to launch. I'm running Adrenaline 19.2.2 drivers. Claymore launches, and then just sits and does nothing. eventually closes. I noticed that AMD settings won't lauch, which I take to be the problem. Did some googling and tried uninstalling the Intel integrated graphics driver. no help. Process is (hooked up to internet)
Reboot in safe mode.
DDU clean restart
install 19.2.2
restart
pixel clock patcher
restart

The computer sees the GPUs  both in device manager and in GPUz, but radeon settings will never launch. I see it in the system tray, but it never opens.

Up to date windows 10 OS. HDMI dummy in one fot he GPUs.  Same setup works on other device.

[MATHReX]

I'm having problems getting Claymore to launch. I'm running Adrenaline 19.2.2 drivers. Claymore launches, and then just sits and does nothing. eventually closes. I noticed that AMD settings won't lauch, which I take to be the problem. Did some googling and tried uninstalling the Intel integrated graphics driver. no help. Process is (hooked up to internet)
Reboot in safe mode.
DDU clean restart
install 19.2.2
restart
pixel clock patcher
restart

The computer sees the GPUs  both in device manager and in GPUz, but radeon settings will never launch. I see it in the system tray, but it never opens.

Up to date windows 10 OS. HDMI dummy in one fot he GPUs.  Same setup works on other device.

Change the graphics display to IGFX in BIOS.
Connect the dummy HDMI to the motherboard slot.
DDU and install the latest version of the drivers. Patch if GPUs are BIOS modded.
Then run the miner.