try to move the china one to the edge of the rig so the fan is facing open air. this will give it the extra cooling it needs. the center of the rig tend to be the hot spot in my experience.
One of my 5700 (Taiwan ) goes upto 104-106°C , thats in Air conditioned room! please do change the Thermal paste/pad to see if that would make any difference.
Same thing happen to me, driving me crazy! I think its a driver issue. I have mixed MSI Saphire 580 5700 Gigabyte and Powercolor. I reinstalled windows, it works better so far.
You may not be out of the woods just yet Panda. Managed to get my 8x Gigabyte 5700XT rig stable for almost 72 hours but the GPU with the highest mem temps finally gave up after 71hrs and 47mins. I've lowered the mem clock on that one. Now the mem is at 82C which hopefully will be good enough. My settings are -42% PL and mem at 1800. This gets me 52 MH/s on the stock BIOS on the 7 cards that seem to be stable. The GPU that crashed I've lowered down to -45% PL and 1750 mem. That one is doing 50.6 MH/s.
@@jlloalv Well if you're doing this in RBE then just double click on 1150 and you will get menu with tCL and tREF and other settings. Or just input this strap FOR MICRON MEMORY ONLY !!! 800900007D040000000000006A09000040020000142619133D00170006060002100000000508080014000000218E06000000010101000101001C1800E02E00000A0A0000B4000000B40000008A206C024C0000000A00040A130300011F2A0C00A500000002F4130FA500000000000000
Excellent video, can't wait to build a rig someday, i'm farming on a couple RX cards past couple weeks, dipping my toes in the coin waters, loving it so far, learning about diff coins and how stuff works, I do have a question, looks like to me atleast, loads of coin crashed couple days ago, anyone know why? thanks
RPM similar thing is happening to me on windows with WiFi.Try uninstaling with DDU,check internet conection than instal drivers with only one gpu and test a bit.If it works add one more gpu,test and repeat. Works for me, but when it happens its almost a day of work :( I think its AMD drivers, they wanna change setting in the middle of minning(sometimes)
Nice! If you decide that the red devil is not needed I can find a place for it 😅😅😅. I am gonna look into the power play setting on that bios editor. I am sure there are ways to reduce the power on the 5700's.
Sure it was the onboard motherboard network, it rarely happens things like this but it does. An alternative to see if is the onboard network card is try simplemining with an wifi usb adapter, if the wifi works then yes is the onboard network card.
Red Panda, my OC setting that are working nicely (and probably room to reduce power a bit more) are in HiveOS Core: 1375 DPM: 2 (doesn’t appear to do much if left blank) VDD/VDDCI/MVDD: 750/750/1275 (the excess under volt is to keep things a little cooler and seems to help a bit) Mem: 930 (pretty much the limit) PL: 105 Get about 1-2% invalid shares on Ethermine Edit: Hashrate: 53.5Mh/s @ 90W (software, on a Sapphire Nitro+ RX 5700 XT)
Its more stable at 56 mh with more voltage. Above 825 to 850 Voltage works for me. It uses a little more power but keeps it stable. In More Power Tools, i set the minimum GFX Voltage to 850. 1432 mhz Clock, Maximum voltage 1050. I use windows. My cards are Gigabyte 5700xt OC rev2, with flashed rev1 2019 Bios. *EDIT* My Fan is set to 75% PWM Minimum and it keeps it stable, i dont know why.
Curious if youd do a video explaining pay outs of pools versus solo. At issue are pps and pps+ pools raking in the huge amounts of block transaction fees while only paying out block rewards. Now with eth and gas prices, yes weve seen a mining payout increase but that's been defi transactions increasing hence block rewards coming faster. Using f2pool, a pps+, and nice hash, a pps, as examples you'll see little to no difference in payouts per h/s. However if f2pool is paying out block transaction fees as pool rewards and with gas prices at their highest levels, there should be a considerable difference. Reason I ask about this is as a miner and blockchain aging, block rewards shrink and block transaction fees take over for the sustainability of the hash rate. Hence then block transaction fees will naturally rise to meet mining costs. E2 and PoS try to lower the costs, both hardware and energy, but basically remove reward for risk, hence why most PoS havent been "successful". Only the rich, and "rich" pools will get transaction fees and payouts, but in a pooled sense if pools keep the transaction fees, then miners (edit: validators when PoS but same basic function) will be getting less and less till the % return for some PoS coins is only 2 to even 10% while the risk is extremely high. Youd be better off buying dow or mid cap stocks for a historical return of 8%, without risk of an new technology. Sorry if it seems a rant, I'm just curious about pools keeping block transaction fees and only paying out block rewards. And your opinion. Seems like I should focus on setting up pools to "borrow" other peoples hash then be an actual miner.
Hi RedPanda! Awesome video series for 5700's. I'm running on a Asus Strix 5700 (3 fans) and, thanks to your bios mod tutorial, I'm able to get 57.25 mh/s stable 24/7. This are OC settings (hope may be usefull also for other guys) Vcore: 800mV, Memory on 1860 mHz, Core at 1350 mHz, powerlimit at zero and fan stable at 80%. With all these I'm stable on 86°C with an average temperature oscillating between 19 and 22°C Luckily now in Italy it's full winter season :P
I had the same problem with my rig, 8 x RX 5700 XT Sapphire Nitro+, motherboard B360-F PRO (MS-7B25), CPU Intel(R) Pentium(R) Gold G5400 CPU @ 3.70GHz, 2 x PSU Chieftec GPU-1050FC, swich LINKSYS LGS124. The whole network put out of service with HiveOS. I did not find the cause of the pane...
Today, I'd the same issue.... two GPU's (1x 1660 TI OC and Red Devil RX 5700 no BIOS flash, just 48MH/s, Temp. fine under 88) on separate Maotherboards. At 15:00 CET (after 36 hours again) both Offline on HiveOS (same time). As I went to check them... ... the fans of the GPUs were off but the CPU was still on. Network orange on and green flashes ... Some steps tried... The only thing that helped was 2x reboot ... works again. For me it is the second exit of the GPUs and the first time that the 1660 Ti OC is not online either..... Confusing
I had similar experience with mixing the ASUS DUAL with other cards and I warned you about when you asked which 5700 to buy, that is where I will try to fish for a possible faulty one or one's.
So I was/am having similar issues with a 5700 XT rig of mine and HiveOS. I found that if the server crashes, the miner crashes and triggers a reboot, or attempting any overlocking changes the rig fails to boot back up into HiveOS even tho there there is power and the network cable is blinking. The solution that I found, annoying that it is, was to just unplug the usb with HiveOS and give it a few seconds, then plug it in and reboot by toggling the power. I think it’s an issue on the motherboard side where it won’t power/recognize it until loading into the bios. Not very helpful for anyone not next to their rigs, but once the rig is stable, it all seems to work fine.
anxious for your next video : ) lowering memory temperature will allow you to use higher overclock. Also might increase hashrate by using a better timing strap
regarding the flooded network traffic, weird man, I was using Asrock H110 pro btc+ and had the same problem. I changed the mobo to asus b250 mining expert when I saw your video and problem was solved. I thought the mobo was the problem, but apparently when seeing this video it is not.
My sappire pulse 570xt on win10/Nicehash/AsrockX570PG4 is really tricky to tune - it regularly crashes the whole system. I'll find a stable voltage/RAM timing that will go for 12 hours then BAM, reboots, totally unstable, start over. Was using NTune but now just Radeon software. Same story on PhoenixMainer and ClaymoreDual (the latter seems easier to get stable). Hashrates start around 51Mh/s but settle to about 45Mh/s in 5 minutes.
@@dennisdowhy wtf, hiveos is linux, its wayyyy more stable than windows and support way more gpus without crashing, the usability its easier and the system consume less power and so many other good things that the reason that you expose...
@@dennisdowhy I will do my best to explain it in English to debate your point 1) Don’t bring that pretentious buls.. that you are a software engineer and blah blah, I’m also one and that means nothing to me, if you can’t validate your point, it does not matter if it comes from Einstein himself it will not be supported. Your title its meaningless if it don’t present facts to the table so it’s irrelevant in the discussion even if it hurts your feelings. 2) in some points it will be stupid to not agree with you, for example, it’s a fact that windows have more variability of programs that help the end user (but that does not mean you cant do it in linux), and some other points I have to recognize that. 3) It’s a fact that Linux it’s a much more stable SO than windows and much more “safety” in terms of security (and yes, I know what would you say about that) otherwise the country’s and all the company’s main servers will not be 98% linux distros. 4) if you don’t like hiveos its understandable but that’s only 1 fish in the see, you have A LOT more options like RAVEOS, simplemining, or even use Debian or ubuntu, fedora etc… 5) When you have electricity problems like my country or a LOT of country’s in the world, windows it’s a BIG problem, its countless the times that I have to manually assist Windows after a power outbreak because windows don’t recover well all the time and in some times you have to even reinstall windows, losing money… 6) Swaping hardware in Windows like replace a bad motherboard it’s a MAJOR pain in the ass and not always go has smooth has linux 7) It’s a lie that you cannot do the same things in linux to the gpus like Undervolting settings, but I have to recognize that you have a lot of software variety (but that can be also a problem much of the time because you have to deal with a lot of information and incompatibility between programs) 8) It happens to a LOT of people that windows upgrades by himself without the permission of the end-user even if you disable in registry and mess with the mining software. 9) much of the security programs also mess a lot of the time with the software 10) setting linux its WAYYYY easier than windows, and you can setup a farm of different hardware in a breezy contrary to windows even if you clone the systems… 11) Linux its so Light that can be use in a flash drive without too much hassle an save in building cost… 12) You have problems in windows like Virtual memory and Windows Versions that has to be all the time revisiting to see witch it’s the best suitable for your rig, or 4g cards that are already presenting problems in windows and with linux its working fine without have to deal with that problem for now… 13) ITS NOT TRUE that its easier the remote management in windows, in fact its so easy and fast in linux that I can do it almost with an Atary and swap mining coins in all my rigs in just seconds, teleconsole, SSH, graphic interface, all that you want… 14) Crashing has a LOT to do with the SO but that is a large conversation also, and not only that, it was not all the time that you could run 13 gpus in windows and has you say it most be 13 identical gpu because if you start to mix in windows you have a world of trouble in front of you, and 19 was much worst also… 15) You ever wonder why almost all the big GPU farms use Linux base systems? Not “small miners” like us that have 6 or 8 rigs, they most have some good reasons don’t you think so? Or you are going to tell me that all that people are users that “could not figure out how to use windows”?. 15) must of the LINUX DISTRO ARE COMPLETELY free, you have to pay the windows license and deal with the company decisions that if they want, they can make windows 10 obsolete in a finger snap. 17) Usability of windows interface is NOT better, its way easier in linux and much more intuitive than windows 16) I could go on just like you say but I think I expose a good amount of proof that debates your points. If you prefer Windows its fine for me but don’t come here with that pretensions claims saying that people that use Linux distros for mining are because they don’t know how to deal with windows, that’s insulting, it’s because we know what is to deal with windows that we choose to use linux….
Everyone knows that eth hash uses Memory for hashing. Just drop the core as far as possible and push the memory as far as possible. And 950mv is fine. You will get better temps and 60mh easy. Stable and power draw comes from the core. So.. ya
Wow, you use high voltages and clocks. I use 1200 core, 850mem, 750 mv core, mvdd 1250, mvddci 750 and get 48.7-50MH/s at ~750-800W @ wall for 6x RX5700. These are stock bios. 6GPU RIG averages ~294.50MH/s -- so that's ~2.71watts per MH, and yours is using ~2.95 watts per MH. ~8.9% power draw increase for just ~7MH/s per card. Need to do some excel to see if that trade off pays off ON AVERAGE, instead of current sky high return. And here i was thinking i need more time to fine tune as i am only getting that ~49MH/s per card on average, or they would have memory overheat issues and some cards run at 51.5-53MH/s while others throttle. I did conservative settings as i did not have time to increase cooling per rig, but maybe it's better to hit that lower power consumption. Sure it is ~11% hash rate increase at "just" 8.9% efficiency loss -- But ultimately this game is about efficiency, not peak rates as much.
Hi Red panda I was facing the same issue with my 5700xt it is because of overclocking. Try to run it on default settings it will work fine without any issues you do not have to change your motherboard.
Good evening Congratulations on the channel and on your work. I would like to ask you that with so many rigs and so much current to load the panel, how does the current panel withstand, have you done something special, or do we need to do something about the current before we deal with a lot so that the switch does not fall, thank you
I have a rig with these same GPUs running on Hive beta version, these GPU won't be recognized in the Non-Beta OS. How did you get it working with the Non-beta?
Even at 750mv core in Hive OS at 58mhs, I'm getting 150w per gpu at the wall! I'm gonna try Windows to see if the power consumption is better, I heard Hive OS Linux drivers may not be as good for these gpus
Nice and stabe like my Asus evo rigs bro! Good job! I think running all the same cards in a rig is helpful. Glad you figured out your issues and its stable.
You almost just lose 1 MH/h per card (and safe more power, i think) and your stable... so well done... be happy :) i also used this mining board at the beginning and also had serveral problems and sent it back... i am rly exited seeing how to use these thermal pads... cya
Mmm Yeah I have a 5700Xt rig with 6 Asus Tuf Evo Cards (yeah the one everyone hates) well im running 780W in the miner, hashrate at 334Mh/s for the 6, not too bad, been running a few days without issues, also my memory temps are between 86-92 depending on the cards. Overall Im satisfied, but yeah your network flood definitaly looks weird
Thank you for showing how to BIOS MOD! I have been trying to get answers the last couple days. You got one big ass LIKE from me Panda THANK YOU! I have to switch my rigs over to HIVE OS. I use a one click right now for the ease of use. I am a beginner and its the easiest to get started. I need to look into HIVE OS. I use windows machines and do not know if that will be an issue! Gotta get researching! THANK YOU AGAIN! -CaptainFrandy
RED PANDA I HAD THE EXACT SAME THING HAPPEN TO ME!! I have an ASUS B250 expert, and I remember that it was a fault in a GPU or a riser, but it would basically DDOS my house hahaha!! I tried opening ports and I tried everything really really annoying.
Will Windows and Nicehash recognize all the GPUs on this Gigabyte motherboard or do some settings be needed. I am beginner and for start I will try windows 10 and nicehash.
all my navi cards i cant use the same clocks i use in windows for mining os thats why i reboot a ton just finding the right clocks but i use raveos cuz i cant get navi working on hive
@@tacoterito22 decided to give hiveOS a try again to see if anything changed, based on ur comment and since its been a few weeks this OS is shit if navi isnt crashing, at settings in every other OS it runs perfect on, polaris is crashing i got cards im running at stock mem just to see if its stable while im reflashing raveos
I could really use some help. I have a small rig with 2x 5700xt and I cannot get it to be stable for more than a day. I did a bios mod on both of them and a very small overclock on memory. My settings are volt 1079 core 1800 and memory 1820. What am I doing wrong???
sounds like it's doing a DOS attack - check for malware. When I started mining, I get a lot more weird incoming traffic from random addresses, not related to mining.
I take it the motherboard was the problem........ Still gotta try the bios mod you did on my 5700's. In hive they are getting in 24 hours 6612 valid shares 102 stale and 3 invalid. Just got to find the time to play with them. LOL Oh yeah, and I got a couple new cards coming....................
I was having troubles with my internet and a rig with only 2 rx5700s in it. I'd turn on the rig and my internet would reset constantly. I put the 5700 rig on a vpn and never had an issue since.. Must have been an isp issue.
Do not buy any Evo duo, xfx thicc 2, msi Evoke, Asus tuf for 5700 or xt all cards have subpar memory cooling. If you want to avoid what happened to rpm watch hardware inbox review of all these cards. But all the info that rpm proves is good.
I think the thermal pad you are using is duplicate Chinese made as if you check thermal grrizly website it only shows max 2 mm minus pad please do check it
I started minig 1 month ago because of you!!! 🤣I'm New în this domain, every thing i know is because of you. i have a 5x 5700xt rig. I use 1400 clock 900vcore, 1844mhz memory running on windows.(284 mhs) But i have around 3% rejected shares. Is it ok, please help! Nu Thanks!
I'm going to guess before I reach your solution in the video and say you had a bad network card that was spitting out bad packets on the network. It can also be caused by Da Boys virus. Well you change the board you change the network card so I guess we'll never really know. Boy am I glad I switched to the H110 motherboard for my build.
my strix RX5700 card bricked everytime I mod the VRAM timing. Anyone got experience with their strix card? Modded on performance mode BIOS. Any suggestion other than Red BIOS editor?
When the video started, because of the crazy noise, my wife just asked: Is this man working in an airport?
try to move the china one to the edge of the rig so the fan is facing open air. this will give it the extra cooling it needs. the center of the rig tend to be the hot spot in my experience.
One of my 5700 (Taiwan ) goes upto 104-106°C
, thats in Air conditioned room! please do change the Thermal paste/pad to see if that would make any difference.
Same thing happen to me, driving me crazy! I think its a driver issue. I have mixed MSI Saphire 580 5700 Gigabyte and Powercolor. I reinstalled windows, it works better so far.
You may not be out of the woods just yet Panda. Managed to get my 8x Gigabyte 5700XT rig stable for almost 72 hours but the GPU with the highest mem temps finally gave up after 71hrs and 47mins. I've lowered the mem clock on that one. Now the mem is at 82C which hopefully will be good enough. My settings are -42% PL and mem at 1800. This gets me 52 MH/s on the stock BIOS on the 7 cards that seem to be stable. The GPU that crashed I've lowered down to -45% PL and 1750 mem. That one is doing 50.6 MH/s.
1.5mm pads for the EVo’s I reckon, 2.0mm too thick I used GELIDS On mine, those minus 8’s might be different
Hey Panda, if you want to get 6+ Mhs from this rig, change in 1550timing tCL to 14 and tREF to 12000 or E02E in strap directly.
May you put an example of the commands pls?
@@jlloalv Well if you're doing this in RBE then just double click on 1150 and you will get menu with tCL and tREF and other settings. Or just input this strap FOR MICRON MEMORY ONLY !!! 800900007D040000000000006A09000040020000142619133D00170006060002100000000508080014000000218E06000000010101000101001C1800E02E00000A0A0000B4000000B40000008A206C024C0000000A00040A130300011F2A0C00A500000002F4130FA500000000000000
@@АндрейЮрашев-т8д Will try this
Андрей Юрашев thx very much, in both memory types in RBE? (I don’t have it at hand right now to check if 1150 is in both or only in one)
@@jlloalv if u just change in rbe in both. If you change full strap, only for micron(mtk...)
Excellent video, can't wait to build a rig someday, i'm farming on a couple RX cards past couple weeks, dipping my toes in the coin waters, loving it so far, learning about diff coins and how stuff works, I do have a question, looks like to me atleast, loads of coin crashed couple days ago, anyone know why? thanks
You can remove the plastic shroud (it traps the hot air) of that xfx card to reduce temps around 7*c
credits to the reddit post~
RPM similar thing is happening to me on windows with WiFi.Try uninstaling with DDU,check internet conection than instal drivers with only one gpu and test a bit.If it works add one more gpu,test and repeat. Works for me, but when it happens its almost a day of work :( I think its AMD drivers, they wanna change setting in the middle of minning(sometimes)
Nice! If you decide that the red devil is not needed I can find a place for it 😅😅😅. I am gonna look into the power play setting on that bios editor. I am sure there are ways to reduce the power on the 5700's.
This rig is the DEVIL that's a whole new issue I never heard off hahah
Sure it was the onboard motherboard network, it rarely happens things like this but it does. An alternative to see if is the onboard network card is try simplemining with an wifi usb adapter, if the wifi works then yes is the onboard network card.
Red Panda, my OC setting that are working nicely (and probably room to reduce power a bit more) are in HiveOS
Core: 1375
DPM: 2 (doesn’t appear to do much if left blank)
VDD/VDDCI/MVDD: 750/750/1275 (the excess under volt is to keep things a little cooler and seems to help a bit)
Mem: 930 (pretty much the limit)
PL: 105
Get about 1-2% invalid shares on Ethermine
Edit: Hashrate: 53.5Mh/s @ 90W (software, on a Sapphire Nitro+ RX 5700 XT)
Is your bios strap modified?
TechisolCorp Not yet. Still running the stock bios since I haven’t had a chance to fiddle with it yet.
@@Fahrenheit353 Nice!
panda with cable management... 2020 is a wierd year lol. love the vid man i want to get either a 57 or 5500 xt soon .
5500xt? Rx580 8gb is better.
@@nikolakarovic5964 is the 5600 better than the 580? cant quite afford the 5700 atm need to mine more to trade up
@@travisstanberry1522 5600xt is good 44mh on eth
Thanks man
Its more stable at 56 mh with more voltage. Above 825 to 850 Voltage works for me. It uses a little more power but keeps it stable. In More Power Tools, i set the minimum GFX Voltage to 850. 1432 mhz Clock, Maximum voltage 1050. I use windows. My cards are Gigabyte 5700xt OC rev2, with flashed rev1 2019 Bios. *EDIT* My Fan is set to 75% PWM Minimum and it keeps it stable, i dont know why.
may i ask you that what temp does your cards stay at on this? especially mem temp
Curious if youd do a video explaining pay outs of pools versus solo. At issue are pps and pps+ pools raking in the huge amounts of block transaction fees while only paying out block rewards. Now with eth and gas prices, yes weve seen a mining payout increase but that's been defi transactions increasing hence block rewards coming faster. Using f2pool, a pps+, and nice hash, a pps, as examples you'll see little to no difference in payouts per h/s. However if f2pool is paying out block transaction fees as pool rewards and with gas prices at their highest levels, there should be a considerable difference. Reason I ask about this is as a miner and blockchain aging, block rewards shrink and block transaction fees take over for the sustainability of the hash rate. Hence then block transaction fees will naturally rise to meet mining costs. E2 and PoS try to lower the costs, both hardware and energy, but basically remove reward for risk, hence why most PoS havent been "successful". Only the rich, and "rich" pools will get transaction fees and payouts, but in a pooled sense if pools keep the transaction fees, then miners (edit: validators when PoS but same basic function) will be getting less and less till the % return for some PoS coins is only 2 to even 10% while the risk is extremely high. Youd be better off buying dow or mid cap stocks for a historical return of 8%, without risk of an new technology. Sorry if it seems a rant, I'm just curious about pools keeping block transaction fees and only paying out block rewards. And your opinion. Seems like I should focus on setting up pools to "borrow" other peoples hash then be an actual miner.
Hi RedPanda! Awesome video series for 5700's. I'm running on a Asus Strix 5700 (3 fans) and, thanks to your bios mod tutorial, I'm able to get 57.25 mh/s stable 24/7.
This are OC settings (hope may be usefull also for other guys) Vcore: 800mV, Memory on 1860 mHz, Core at 1350 mHz, powerlimit at zero and fan stable at 80%. With all these I'm stable on 86°C with an average temperature oscillating between 19 and 22°C
Luckily now in Italy it's full winter season :P
I had the same problem with my rig, 8 x RX 5700 XT Sapphire Nitro+, motherboard B360-F PRO (MS-7B25), CPU Intel(R) Pentium(R) Gold G5400 CPU @ 3.70GHz, 2 x PSU Chieftec GPU-1050FC, swich LINKSYS LGS124. The whole network put out of service with HiveOS. I did not find the cause of the pane...
It would be mince to know how much power consumes each card from this particular wideo :)
I have three 5700 card, the average speed is 54.5 - 55.6 mh, 118-121 w, and are stable. I think this is a good performance.
Did you update the bios on the b250 mining expert?
Today, I'd the same issue.... two GPU's (1x 1660 TI OC and Red Devil RX 5700 no BIOS flash, just 48MH/s, Temp. fine under 88) on separate Maotherboards. At 15:00 CET (after 36 hours again) both Offline on HiveOS (same time). As I went to check them...
... the fans of the GPUs were off but the CPU was still on. Network orange on and green flashes ... Some steps tried...
The only thing that helped was 2x reboot ...
works again.
For me it is the second exit of the GPUs and the first time that the 1660 Ti OC is not online either..... Confusing
I had similar experience with mixing the ASUS DUAL with other cards and I warned you about when you asked which 5700 to buy, that is where I will try to fish for a possible faulty one or one's.
Nice work. How many Watt in wall?
I think problem is in very high temperature of memory of the cards.
So I was/am having similar issues with a 5700 XT rig of mine and HiveOS. I found that if the server crashes, the miner crashes and triggers a reboot, or attempting any overlocking changes the rig fails to boot back up into HiveOS even tho there there is power and the network cable is blinking. The solution that I found, annoying that it is, was to just unplug the usb with HiveOS and give it a few seconds, then plug it in and reboot by toggling the power. I think it’s an issue on the motherboard side where it won’t power/recognize it until loading into the bios. Not very helpful for anyone not next to their rigs, but once the rig is stable, it all seems to work fine.
I have the same issue, and im about to try installing hiveos on a hdd drive instead of usb
You should let me know if this solves the issue. I might have to try it out.
Is there any difference between a rx 5700 xt and a rx 5700 for mining?
anxious for your next video : )
lowering memory temperature will allow you to use higher overclock. Also might increase hashrate by using a better timing strap
regarding the flooded network traffic, weird man, I was using Asrock H110 pro btc+ and had the same problem. I changed the mobo to asus b250 mining expert when I saw your video and problem was solved. I thought the mobo was the problem, but apparently when seeing this video it is not.
My sappire pulse 570xt on win10/Nicehash/AsrockX570PG4 is really tricky to tune - it regularly crashes the whole system. I'll find a stable voltage/RAM timing that will go for 12 hours then BAM, reboots, totally unstable, start over. Was using NTune but now just Radeon software. Same story on PhoenixMainer and ClaymoreDual (the latter seems easier to get stable). Hashrates start around 51Mh/s but settle to about 45Mh/s in 5 minutes.
Im in the middle of building a 5700XT Red Devil rig. What is your feeling on the future of mining with ETH 2.0 in the horizon?
With AMD, google "amd August drivers" August drivers will stop blue screen crashes, and reset issues.
Red Panda Ddoxed with his motherboard 😁
Nikola sto neotvoris svoj kanal daleko vise iskustva imas nego ovi pocetnici :D
What’s the difference between hive os and Windows 10
@@dennisdowhy wtf, hiveos is linux, its wayyyy more stable than windows and support way more gpus without crashing, the usability its easier and the system consume less power and so many other good things that the reason that you expose...
@@dennisdowhy Sure prove it and then I will believe you, because my experience say other things...
@@dennisdowhy I will do my best to explain it in English to debate your point
1) Don’t bring that pretentious buls.. that you are a software engineer and blah blah, I’m also one and that means nothing to me, if you can’t validate your point, it does not matter if it comes from Einstein himself it will not be supported. Your title its meaningless if it don’t present facts to the table so it’s irrelevant in the discussion even if it hurts your feelings.
2) in some points it will be stupid to not agree with you, for example, it’s a fact that windows have more variability of programs that help the end user (but that does not mean you cant do it in linux), and some other points I have to recognize that.
3) It’s a fact that Linux it’s a much more stable SO than windows and much more “safety” in terms of security (and yes, I know what would you say about that) otherwise the country’s and all the company’s main servers will not be 98% linux distros.
4) if you don’t like hiveos its understandable but that’s only 1 fish in the see, you have A LOT more options like RAVEOS, simplemining, or even use Debian or ubuntu, fedora etc…
5) When you have electricity problems like my country or a LOT of country’s in the world, windows it’s a BIG problem, its countless the times that I have to manually assist Windows after a power outbreak because windows don’t recover well all the time and in some times you have to even reinstall windows, losing money…
6) Swaping hardware in Windows like replace a bad motherboard it’s a MAJOR pain in the ass and not always go has smooth has linux
7) It’s a lie that you cannot do the same things in linux to the gpus like Undervolting settings, but I have to recognize that you have a lot of software variety (but that can be also a problem much of the time because you have to deal with a lot of information and incompatibility between programs)
8) It happens to a LOT of people that windows upgrades by himself without the permission of the end-user even if you disable in registry and mess with the mining software.
9) much of the security programs also mess a lot of the time with the software
10) setting linux its WAYYYY easier than windows, and you can setup a farm of different hardware in a breezy contrary to windows even if you clone the systems…
11) Linux its so Light that can be use in a flash drive without too much hassle an save in building cost…
12) You have problems in windows like Virtual memory and Windows Versions that has to be all the time revisiting to see witch it’s the best suitable for your rig, or 4g cards that are already presenting problems in windows and with linux its working fine without have to deal with that problem for now…
13) ITS NOT TRUE that its easier the remote management in windows, in fact its so easy and fast in linux that I can do it almost with an Atary and swap mining coins in all my rigs in just seconds, teleconsole, SSH, graphic interface, all that you want…
14) Crashing has a LOT to do with the SO but that is a large conversation also, and not only that, it was not all the time that you could run 13 gpus in windows and has you say it most be 13 identical gpu because if you start to mix in windows you have a world of trouble in front of you, and 19 was much worst also…
15) You ever wonder why almost all the big GPU farms use Linux base systems? Not “small miners” like us that have 6 or 8 rigs, they most have some good reasons don’t you think so? Or you are going to tell me that all that people are users that “could not figure out how to use windows”?.
15) must of the LINUX DISTRO ARE COMPLETELY free, you have to pay the windows license and deal with the company decisions that if they want, they can make windows 10 obsolete in a finger snap.
17) Usability of windows interface is NOT better, its way easier in linux and much more intuitive than windows
16) I could go on just like you say but I think I expose a good amount of proof that debates your points. If you prefer Windows its fine for me but don’t come here with that pretensions claims saying that people that use Linux distros for mining are because they don’t know how to deal with windows, that’s insulting, it’s because we know what is to deal with windows that we choose to use linux….
Everyone knows that eth hash uses Memory for hashing. Just drop the core as far as possible and push the memory as far as possible. And 950mv is fine. You will get better temps and 60mh easy. Stable and power draw comes from the core. So.. ya
can you suggest how to supply power to those gpus and risers to get a clean rig?
Looks pretty! You can clock them to 60Mh+
Wow, you use high voltages and clocks. I use 1200 core, 850mem, 750 mv core, mvdd 1250, mvddci 750 and get 48.7-50MH/s at ~750-800W @ wall for 6x RX5700.
These are stock bios. 6GPU RIG averages ~294.50MH/s -- so that's ~2.71watts per MH, and yours is using ~2.95 watts per MH. ~8.9% power draw increase for just ~7MH/s per card.
Need to do some excel to see if that trade off pays off ON AVERAGE, instead of current sky high return.
And here i was thinking i need more time to fine tune as i am only getting that ~49MH/s per card on average, or they would have memory overheat issues and some cards run at 51.5-53MH/s while others throttle. I did conservative settings as i did not have time to increase cooling per rig, but maybe it's better to hit that lower power consumption.
Sure it is ~11% hash rate increase at "just" 8.9% efficiency loss -- But ultimately this game is about efficiency, not peak rates as much.
Do you have any on the samsung ram? I have 2/8 on my 8 card rig that are Samsung and I just cannot keep them on. I'm pretty sure its the overclocks.
Hi Red panda I was facing the same issue with my 5700xt it is because of overclocking. Try to run it on default settings it will work fine without any issues you do not have to change your motherboard.
How is he powering the gpus and risers with only 6 PCIE outlets? Is he using molex to power the risers?
Why are gpu 0 and gpu 2 using more power than the rest?
Good evening Congratulations on the channel and on your work.
I would like to ask you that with so many rigs and so much current to load the panel, how does the current panel withstand, have you done something special, or do we need to do something about the current before we deal with a lot so that the switch does not fall, thank you
do you always run you gpu fans at 100%? doesnt that burn them out faster?
I always have that issue on Sat 3pm-7pm it’s a internet provider issue. What service provider are you using?
I have a rig with these same GPUs running on Hive beta version, these GPU won't be recognized in the Non-Beta OS. How did you get it working with the Non-beta?
Even at 750mv core in Hive OS at 58mhs, I'm getting 150w per gpu at the wall! I'm gonna try Windows to see if the power consumption is better, I heard Hive OS Linux drivers may not be as good for these gpus
Hi Panda,
what kind of miner do you use?
Thanks in advance.
Teamredminer
how did you get the rig to turn on from the PSU power switch alone instead of the assigned power button like a PC?
Most motherboards have a bios power setting for this.
When you show temperature inside your mine, could you switch the temperature from F to C for people from Europe?
Do you turn off all the LEDs they do take up small amount of power little but
Nice and stabe like my Asus evo rigs bro! Good job! I think running all the same cards in a rig is helpful. Glad you figured out your issues and its stable.
You almost just lose 1 MH/h per card (and safe more power, i think) and your stable... so well done... be happy :) i also used this mining board at the beginning and also had serveral problems and sent it back... i am rly exited seeing how to use these thermal pads... cya
Mmm Yeah I have a 5700Xt rig with 6 Asus Tuf Evo Cards (yeah the one everyone hates) well im running 780W in the miner, hashrate at 334Mh/s for the 6, not too bad, been running a few days without issues, also my memory temps are between 86-92 depending on the cards. Overall Im satisfied, but yeah your network flood definitaly looks weird
Thank you for showing how to BIOS MOD! I have been trying to get answers the last couple days. You got one big ass LIKE from me Panda THANK YOU!
I have to switch my rigs over to HIVE OS. I use a one click right now for the ease of use. I am a beginner and its the easiest to get started. I need to look into HIVE OS.
I use windows machines and do not know if that will be an issue! Gotta get researching! THANK YOU AGAIN!
-CaptainFrandy
How did you manage 53 mh/s stock? Was sitting at 51 then restarted and I'm at 52
I want ask you. Did you got coinbase when eth send to wallet to coinbase?
Use 1 device per VLAN id so less local traffic. And do you use VPN?
RED PANDA I HAD THE EXACT SAME THING HAPPEN TO ME!!
I have an ASUS B250 expert, and I remember that it was a fault in a GPU or a riser, but it would basically DDOS my house hahaha!! I tried opening ports and I tried everything really really annoying.
Will Windows and Nicehash recognize all the GPUs on this Gigabyte motherboard or do some settings be needed. I am beginner and for start I will try windows 10 and nicehash.
How did you reduce Memory voltage to 800? Please i need to know this.
by lowering it. duh
@@sniperpronerfmods9811 If not troll, no mem voltage control in msi afterburner or wattman and phoenix miner command doesnt work. Don't know how.
@@eugene4637 use claymore bat file mvddc
How can I or is it possible to get Windows 10 to see more than 8 gpus? I have not uncovered anything regarding this. Thanks for your time.
yay
Hai red panda, how much total watt on the wall u got for this rig?
all my navi cards i cant use the same clocks i use in windows for mining os
thats why i reboot a ton just finding the right clocks
but i use raveos cuz i cant get navi working on hive
you have to use Hiveos BETA version for the rx 5700xt
@@tacoterito22 neither worked
thats why im using raveos
@@tacoterito22 decided to give hiveOS a try again to see if anything changed, based on ur comment and since its been a few weeks
this OS is shit
if navi isnt crashing, at settings in every other OS it runs perfect on, polaris is crashing
i got cards im running at stock mem just to see if its stable while im reflashing raveos
Hi. 6 x 5700 how many watts it draws from the wall? Ty.
I could really use some help. I have a small rig with 2x 5700xt and I cannot get it to be stable for more than a day. I did a bios mod on both of them and a very small overclock on memory. My settings are volt 1079 core 1800 and memory 1820. What am I doing wrong???
So are you using windows directly on the rig to do the vbios mod? Or can you do it from a separate desktop?
Yep separate windows machine to use red bios editor and chrome to access hiveos
@@RedPandaMining thank you!!!
sounds like it's doing a DOS attack - check for malware. When I started mining, I get a lot more weird incoming traffic from random addresses, not related to mining.
Great vid man. What do you mean when you say rx 5700 cards are “end of life”?
They dont make them anymore, only 5700 xt now
Please show us how this rig performed.
Hi Red panda mining.
Nice video like always.
It's been 6 month now. Any feedback about the long term use of this kind of "hot OC".
THANKS by advance.
Can the new BIOS mod be flashed even when cudo miner is used?.
I take it the motherboard was the problem........ Still gotta try the bios mod you did on my 5700's. In hive they are getting in 24 hours 6612 valid shares 102 stale and 3 invalid. Just got to find the time to play with them. LOL Oh yeah, and I got a couple new cards coming....................
Can someone please guide me...what OS is better..I’m building a rig
Hi, which is better brand for 5700xt?
Why do you use Ethernet connection over just usb wifi?
what happened to the 5700s? why cant you buy them new anymore?
hi bro i have 5700 xt and is no stable is same solution? i have 2 red devils 1 msi and 1 xfx alls 5700xt
hive os is TOP mining system!!!! ;)
how do you consume electricity on the wall with rx5700xt?
Red Panda bro with that hot Chinese card just move it to the end where the fans aren’t sucking in so much heat
I was having troubles with my internet and a rig with only 2 rx5700s in it. I'd turn on the rig and my internet would reset constantly. I put the 5700 rig on a vpn and never had an issue since.. Must have been an isp issue.
You have EVGA made risers?
How do you fit 6 dual 8 pin gpu's on a single psu?
Do not buy any Evo duo, xfx thicc 2, msi Evoke, Asus tuf for 5700 or xt all cards have subpar memory cooling. If you want to avoid what happened to rpm watch hardware inbox review of all these cards. But all the info that rpm proves is good.
I think the thermal pad you are using is duplicate Chinese made as if you check thermal grrizly website it only shows max 2 mm minus pad please do check it
How many w power consumption at wall?
I started minig 1 month ago because of you!!! 🤣I'm New în this domain, every thing i know is because of you. i have a 5x 5700xt rig. I use 1400 clock 900vcore, 1844mhz memory running on windows.(284 mhs) But i have around 3% rejected shares. Is it ok, please help! Nu Thanks!
Is teamredminer better than phoenix?
для тушек используй материнские платы гигабайт финтеч б250 на 12 карт, самые беспроблемные материнки, проверено годами!
Had this with one of mine rigs - flooding the network - I changed the risers and it stopped crashing my network.
hi.use rx 5700xt sapphire nitro + , asus dual have not good cooler
why are 2 of those such high watt??
I look forwards to your videos every day, keep it up!
whats with the mega drop in profits past 3 days? is it my rigs?
Check out the ethereum block rewards they dropped in half pretty much
@@RedPandaMining figures as soon as i expand... :(
I'm going to guess before I reach your solution in the video and say you had a bad network card that was spitting out bad packets on the network. It can also be caused by Da Boys virus. Well you change the board you change the network card so I guess we'll never really know. Boy am I glad I switched to the H110 motherboard for my build.
Hey everybody, can I run one 5700 in a regular desktop box without it running hot? Thanks.
my strix RX5700 card bricked everytime I mod the VRAM timing. Anyone got experience with their strix card? Modded on performance mode BIOS. Any suggestion other than Red BIOS editor?
rom flashing field error ?