My second ZEC + XMR+ ETH thread builds info links thoughts and photos.

[LoneRangir]

Meanwhile.... I have been waiting for AMD's ZEN CPU.... so it has been launched as "RYZEN" -- anyone has gotten one yet?

Are you going to mine on it or it is just a powerful CPU?

zen is suppose to compete with intel on everything from now on
so its going to be a powerful cpu
 right now the ryzen is a 4-8 core cpu with 8-16 thread starting at 3.4ghz with 20mb L2+L3 cache

been a long time since I built an amd pc.

but it is still running at pne of my friends places

I just bought a used GA-990FXA-UD7 Mobo/ AMD CPU pair. It has 6 PCIe slots, was thinking of using risers. It also has the correct spacing to just run 4 cards straight up. Does anyone have practical experience with this board? I may just keep it simple and make it easy on myself after reading the last few pages 

[philipma1957]

Phil,

Does the Biostar Z170GT7 (or the Z170 chipset) support up to 4 GPUs or 4 physical graphics cards?
Most reports I read say you can't run 5 cards on this board. I have a 295x2 (dual-GPU) card, so wondering if I can stick 4 cards / 5 GPUs and get it to work on this board.

Appreciate any insight on this.

my last 2 card gpu was a power color red devil 7990  I had 3 of them  one was an rma the other two ran one card card in each one and one decent card.

the boards would see them as four cards  but they would use six slots vs eight slots.

so based on my hd 7990's I would say it won't work.

I have seen a few people say they get 5 cards to run on a z170.

I gave up trying.

now in your case if the board lets the dual run with 2 other cards  they would be a 4 card setup that uses 3 cards and  7 slots so it would be extra room for cooling.

[QuintLeo]

I just bought a used GA-990FXA-UD7 Mobo/ AMD CPU pair. It has 6 PCIe slots, was thinking of using risers. It also has the correct spacing to just run 4 cards straight up. Does anyone have practical experience with this board? I may just keep it simple and make it easy on myself after reading the last few pages 

 Back in the day that was considered to be *THE* go-to board for a lot of Litecoin miners.

 I wish Gigabyte still made a -UD7 varient of at at least one of their boards.

[TheRider]

Phil,

Does the Biostar Z170GT7 (or the Z170 chipset) support up to 4 GPUs or 4 physical graphics cards?
Most reports I read say you can't run 5 cards on this board. I have a 295x2 (dual-GPU) card, so wondering if I can stick 4 cards / 5 GPUs and get it to work on this board.

Appreciate any insight on this.

my last 2 card gpu was a power color red devil 7990  I had 3 of them  one was an rma the other two ran one card card in each one and one decent card.

the boards would see them as four cards  but they would use six slots vs eight slots.

so based on my hd 7990's I would say it won't work.

I have seen a few people say they get 5 cards to run on a z170.

I gave up trying.

now in your case if the board lets the dual run with 2 other cards  they would be a 4 card setup that uses 3 cards and  7 slots so it would be extra room for cooling.

Thanks. I think I will try it with the dual gpu card and 2 other cards in the first 3 slots, make sure they're all recognized and mining, and then I might add a fourth card (5th gpu) in the fourth slot to see if it works that way.

[QuintLeo]

Meanwhile.... I have been waiting for AMD's ZEN CPU.... so it has been launched as "RYZEN" -- anyone has gotten one yet?

 2017 - Zen has NOT been launched yet.

 The new "A12" and related series smaller APUs are basically a Godavari refresh on the 28nm node with some minor additional optimisations, and OEM only - though they ARE AM4 socket so you can find motherboards in some OEM machines that can be upgraded to ZEN when they start showing up.

[vg54dett]

Citronick , I have a question again : as your are a usual user of ethman , could tell me how is working the batch file that ethman launch if miner as a problem ?

The bat is called "sample" and include :
"@ECHO OFF

ECHO This is sample batch file that can do something useful if rig has problems
ECHO Rig name: %1
TIMEOUT /T 5"

I guess we have to modify it in order to get result but I don't know what I can put inside this bat ?
Any advice ?

[vg54dett]

Do someone know if our modded gpus (1500 strap) are still efficient for gaming ? or do we loose lot of performance when playing game with a modded bios ?
(it's about a home computer mining when unused)

[wedatin]

Phil,

Does the Biostar Z170GT7 (or the Z170 chipset) support up to 4 GPUs or 4 physical graphics cards?
Most reports I read say you can't run 5 cards on this board. I have a 295x2 (dual-GPU) card, so wondering if I can stick 4 cards / 5 GPUs and get it to work on this board.

Appreciate any insight on this.

my last 2 card gpu was a power color red devil 7990  I had 3 of them  one was an rma the other two ran one card card in each one and one decent card.

the boards would see them as four cards  but they would use six slots vs eight slots.

so based on my hd 7990's I would say it won't work.

I have seen a few people say they get 5 cards to run on a z170.

I gave up trying.

now in your case if the board lets the dual run with 2 other cards  they would be a 4 card setup that uses 3 cards and  7 slots so it would be extra room for cooling.

I do not use the Z170. But I have some old MB such as ASrock H81/H61. I can run 4X7970 + 2X7990 or 4XR9 390 + 4X7990 without any problem. There are 8 GPU in total.

[philipma1957]

Phil,

Does the Biostar Z170GT7 (or the Z170 chipset) support up to 4 GPUs or 4 physical graphics cards?
Most reports I read say you can't run 5 cards on this board. I have a 295x2 (dual-GPU) card, so wondering if I can stick 4 cards / 5 GPUs and get it to work on this board.

Appreciate any insight on this.

my last 2 card gpu was a power color red devil 7990  I had 3 of them  one was an rma the other two ran one card card in each one and one decent card.

the boards would see them as four cards  but they would use six slots vs eight slots.

so based on my hd 7990's I would say it won't work.

I have seen a few people say they get 5 cards to run on a z170.

I gave up trying.

now in your case if the board lets the dual run with 2 other cards  they would be a 4 card setup that uses 3 cards and  7 slots so it would be extra room for cooling.

I do not use the Z170. But I have some old MB such as ASrock H81/H61. I can run 4X7970 + 2X7990 or 4XR9 390 + 4X7990 without any problem. There are 8 GPU in total.

We know that older chipsets will more gpus, but  I would love to see a running example of   4x R9 390 with 4 x 7990   as that is 12 gpus not 8

[ps_jb]

Phil and all,

I spent yesterday an hour reading on Pandaminer on Russian forum. Actually except size and complete integrated solution there are no advantages compared to GPU made rig.

They also noted that copper heat sinks are state-of-the-art.

Waiting to read your review!

[xleejohnx]

Do someone know if our modded gpus (1500 strap) are still efficient for gaming ? or do we loose lot of performance when playing game with a modded bios ?
(it's about a home computer mining when unused)

a modded bios is just a like overclocking with software
the 1500 strap moves the timing of the memory in closer
so all in all youll have better performance when gaming

[Seast]

Phil,

Does the Biostar Z170GT7 (or the Z170 chipset) support up to 4 GPUs or 4 physical graphics cards?
Most reports I read say you can't run 5 cards on this board. I have a 295x2 (dual-GPU) card, so wondering if I can stick 4 cards / 5 GPUs and get it to work on this board.

Appreciate any insight on this.

my last 2 card gpu was a power color red devil 7990  I had 3 of them  one was an rma the other two ran one card card in each one and one decent card.

the boards would see them as four cards  but they would use six slots vs eight slots.

so based on my hd 7990's I would say it won't work.

I have seen a few people say they get 5 cards to run on a z170.

I gave up trying.

now in your case if the board lets the dual run with 2 other cards  they would be a 4 card setup that uses 3 cards and  7 slots so it would be extra room for cooling.

I do not use the Z170. But I have some old MB such as ASrock H81/H61. I can run 4X7970 + 2X7990 or 4XR9 390 + 4X7990 without any problem. There are 8 GPU in total.

We know that older chipsets will more gpus, but  I would love to see a running example of   4x R9 390 with 4 x 7990   as that is 12 gpus not 8

4XR9 390 + 4X7990 was a typo He said 8 GPU in total.

I tried to run 9 GPU, but never succeed.

[citronick]

Citronick , I have a question again : as your are a usual user of ethman , could tell me how is working the batch file that ethman launch if miner as a problem ?

The bat is called "sample" and include :
"@ECHO OFF

ECHO This is sample batch file that can do something useful if rig has problems
ECHO Rig name: %1
TIMEOUT /T 5"

I guess we have to modify it in order to get result but I don't know what I can put inside this bat ?
Any advice ?

If any of my rigs has issues, frankly I am not interested in the why and the details - I even turn the log file off - the last thing on my mind is investigating why it rebooted... because my farm is offsite, I just want the rig to be rebooted quickly. In Windows vs LINUX, it's not secret that Windows bloatware and memory leaks are usually the main culprit in stability and performance of a Windows environment. So reboot is the only way to clear memory stacks and get Windows to start fresh.

In Ethman, it can track how many times your rig rebooted since last Ethman launch. I only keep an eye on this counter. If I see too many reboots then I will babysit the particular rig and see what goes on with the log file turned on - anything out of the ordinary, I will just post and ask Claymore about it -- most likely a Claymore thing eventually rather than a Windows thing. Otherwise, I dont expect it to reboot at all. All my rigs can stay hashing online for very long time, ie. 2-3 weeks 24x7.

Some bad rigs constantly rebooting - usually are PSU, power, temperature, faulty riser issues... sometimes bad Crimson drivers .... or HW errors if you really unlucky...  so almost nothing to do with OS.

This reboot feature with watchdog is very key feature in Ethman - worth every cent in dev fee this Ethman.
 
- r 1 and reboot.bat with a single command line "shutdown /r /t 5 /f"

Please note you are using shutdown command where /r is reboot with 5 seconds countdown and /f force reboot

Extracted from Claymore's README file.
========================================================
-r   Restart miner mode. "-r 0" (default) - restart miner if something wrong with GPU. "-r -1" - disable automatic restarting. -r >20 - restart miner if something    wrong with GPU or by timer. For example, "-r 60" - restart miner every hour or when some GPU failed.
   "-r 1" closes miner and execute "reboot.bat" file ("reboot.bash" or "reboot.sh" for Linux version) in the miner directory (if exists) if some GPU failed.
   So you can create "reboot.bat" file and perform some actions, for example, reboot system if you put this line there: "shutdown /r /t 5 /f".
========================================================

[GabryRox]

Do someone know if our modded gpus (1500 strap) are still efficient for gaming ? or do we loose lot of performance when playing game with a modded bios ?
(it's about a home computer mining when unused)

a modded bios is just a like overclocking with software
the 1500 strap moves the timing of the memory in closer
so all in all youll have better performance when gaming

This might be a stupid question, but if 1500-strapping improved both mining & gaming performance/efficiency, then why don't they come that way from the factory lol?  I mean, those 2 uses cover like 99% of uses for these high end GPUs, so why make customers jump through hoops to get the best performance?

[xleejohnx]

Do someone know if our modded gpus (1500 strap) are still efficient for gaming ? or do we loose lot of performance when playing game with a modded bios ?
(it's about a home computer mining when unused)

a modded bios is just a like overclocking with software
the 1500 strap moves the timing of the memory in closer
so all in all youll have better performance when gaming

This might be a stupid question, but if 1500-strapping improved both mining & gaming performance/efficiency, then why don't they come that way from the factory lol?  I mean, those 2 uses cover like 99% of uses for these high end GPUs, so why make customers jump through hoops to get the best performance?

its marketing
you dont make it perfect so you can sell a newer model for more money

also changing the timing does stress the gpu memory more. under normal loads like gaming it doesnt hurt the gpu as bad as mining

[vg54dett]

Citronick , I have a question again : as your are a usual user of ethman , could tell me how is working the batch file that ethman launch if miner as a problem ?

The bat is called "sample" and include :
"@ECHO OFF

ECHO This is sample batch file that can do something useful if rig has problems
ECHO Rig name: %1
TIMEOUT /T 5"

I guess we have to modify it in order to get result but I don't know what I can put inside this bat ?
Any advice ?

If any of my rigs has issues, frankly I am not interested in the why and the details - I even turn the log file off - the last thing on my mind is investigating why it rebooted... because my farm is offsite, I just want the rig to be rebooted quickly. In Windows vs LINUX, it's not secret that Windows bloatware and memory leaks are usually the main culprit in stability and performance of a Windows environment. So reboot is the only way to clear memory stacks and get Windows to start fresh.

In Ethman, it can track how many times your rig rebooted since last Ethman launch. I only keep an eye on this counter. If I see too many reboots then I will babysit the particular rig and see what goes on with the log file turned on - anything out of the ordinary, I will just post and ask Claymore about it -- most likely a Claymore thing eventually rather than a Windows thing. Otherwise, I dont expect it to reboot at all. All my rigs can stay hashing online for very long time, ie. 2-3 weeks 24x7.

Some bad rigs constantly rebooting - usually are PSU, power, temperature, faulty riser issues... sometimes bad Crimson drivers .... or HW errors if you really unlucky...  so almost nothing to do with OS.

This reboot feature with watchdog is very key feature in Ethman - worth every cent in dev fee this Ethman.
 
- r 1 and reboot.bat with a single command line "shutdown /r /t 5 /f"

Please note you are using shutdown command where /r is reboot with 5 seconds countdown and /f force reboot

Extracted from Claymore's README file.
========================================================
-r   Restart miner mode. "-r 0" (default) - restart miner if something wrong with GPU. "-r -1" - disable automatic restarting. -r >20 - restart miner if something    wrong with GPU or by timer. For example, "-r 60" - restart miner every hour or when some GPU failed.
   "-r 1" closes miner and execute "reboot.bat" file ("reboot.bash" or "reboot.sh" for Linux version) in the miner directory (if exists) if some GPU failed.
   So you can create "reboot.bat" file and perform some actions, for example, reboot system if you put this line there: "shutdown /r /t 5 /f".
========================================================

Thank you very much for the informations.
I'm not sure to understand correctly (maybe due to translations)

When I use -r 0 that restart miner if something wrong with GPU , it makes restart miner as soon as a gpu hang. (the gpu show 0 mh/s)
My problem is that I always get blue screen when miner restart after a gpu hang.
If I want to get all gpu work again, I have to reboot windows.
Of course, a blue screen leads to a reboot.
So what I need is a way to start claymore again after an unexpected windows reboot.
If I understand correctly what you said , the reboot.bat with shutdown /r /t 5 /f , associated with -r 1 would do the trick ? or did I miss something ?
If a gpu hang, so windows will reboot?
Where do we put this reboot.bat ? with ethman or with claymore ?
How do we have claymore launched automatically after windows reboot ?
Sorry if it sounds stupid .. and thank you again ! your advices are precious.

[vg54dett]

Do someone know if our modded gpus (1500 strap) are still efficient for gaming ? or do we loose lot of performance when playing game with a modded bios ?
(it's about a home computer mining when unused)

a modded bios is just a like overclocking with software
the 1500 strap moves the timing of the memory in closer
so all in all youll have better performance when gaming

great news, thank you for confirming that point !

[citronick]

vg54dett - bluescreen will need hard reset - and the reboot batch file will not work on your scenario.

I think you need to solve this bluescreen issue separately first. I suspect, GPU driver or RAM related issues. What is your rig config and setup?

In basic setup, reboot.bat need to be saved in same folder as Claymore miner.

In your file manager browser, type "shell:startup" and hit enter. In this folder, you can put a batch file to run Claymore miner or any other app. When Windows boots up, it will run all programs and batch files in this startup folder.

[GabryRox]

Citronick... can I ask your recommendation on when to run that 6xGPU mod app?  pre or post attempted install of GPU #5?  Meaning, should I run it now with only the 4 working GPUs plugged into my rig or, attempt to install the 5th again (getting the error 43) and THEN run this mod app?  Thanks  

Put in all 5 OR all 6 GPUs.

Install DDU and perform clean DDU in safe mode (Download latest here http://www.guru3d.com/files-details/display-driver-uninstaller-download.html)

Install Crimson 16.11.x (avoid the Relive 16.12.x)

Ensure Device Manager sees them all (you may need to seat and reseat the risers etc)

If the last GPU give you "Error 43" - run the 6GPU utility in Admin mode - reboot.

Good luck!

Thanks to Citronick, I finally got the 5th GPU working on rig #2!  The "funny" thing is that I was trying it before I saw this post by you, and it would not work. Looked like it was writing to the registry but that error 43 still showed up.  Finally, I quit trying, came back and saw your post.  Even though I already knew to run that mod as Admin, I had somehow forgotten to do that on all my attempts.  So, I went back, ran the 6xGPU as admin and amazingly, it worked immediately after rebooting. Now have all 5 cards hashing about 175 S/s per, sucking about 727W at max.

Note that I have #5 plugged in the the last x1 PCIe slot.  I didn't wanna mess with trying the x16 slot on this attempt since i have a feeling that there is a different issue with that slot.  But, instead of taking a chance and buying another GPU outright, I plan to pull 1 off my other rig and test it in that x16 slot.  I am not really optimistic that this 6xGPU mod will have fixed the issue with the x16 slot not working, since I have never even been able to get this rig to boot with the riser plugged into this slot, but I suppose it's at least worth a shot.  Thanks again to Citronick for helping me get past the 4-GPU barrier!

[philipma1957]

Citronick... can I ask your recommendation on when to run that 6xGPU mod app?  pre or post attempted install of GPU #5?  Meaning, should I run it now with only the 4 working GPUs plugged into my rig or, attempt to install the 5th again (getting the error 43) and THEN run this mod app?  Thanks  

Put in all 5 OR all 6 GPUs.

Install DDU and perform clean DDU in safe mode (Download latest here http://www.guru3d.com/files-details/display-driver-uninstaller-download.html)

Install Crimson 16.11.x (avoid the Relive 16.12.x)

Ensure Device Manager sees them all (you may need to seat and reseat the risers etc)

If the last GPU give you "Error 43" - run the 6GPU utility in Admin mode - reboot.

Good luck!

Thanks to Citronick, I finally got the 5th GPU working on rig #2!  The "funny" thing is that I was trying it before I saw this post by you, and it would not work. Looked like it was writing to the registry but that error 43 still showed up.  Finally, I quit trying, came back and saw your post.  Even though I already knew to run that mod as Admin, I had somehow forgotten to do that on all my attempts.  So, I went back, ran the 6xGPU as admin and amazingly, it worked immediately after rebooting. Now have all 5 cards hashing about 175 S/s per, sucking about 727W at max.

Note that I have #5 plugged in the the last x1 PCIe slot.  I didn't wanna mess with trying the x16 slot on this attempt since i have a feeling that there is a different issue with that slot.  But, instead of taking a chance and buying another GPU outright, I plan to pull 1 off my other rig and test it in that x16 slot.  I am not really optimistic that this 6xGPU mod will have fixed the issue with the x16 slot not working, since I have never even been able to get this rig to boot with the riser plugged into this slot, but I suppose it's at least worth a shot.  Thanks again to Citronick for helping me get past the 4-GPU barrier!

Nice  maybe  you can do 5 + 5 on the two rigs.

I just got this pc


 http://store.hp.com/us/en/PDPStdView?catalogId=10051&urlLangId=-1&langId=-1&productId=1501157&storeId=10151


mostly because I wanted to see the rx 480 inside it.  same size as my rx 460 visiontek

photos coming up

shortest one in this photo


here is the visiontek rx 460 well under 8 inches


the hp rx 480 is shorter by about ½ inch