I will re-install win 10 one last time and try the 6 card one rig ... if it fails it'll be 3 GPU per MOBO.
Tired of loosing hours of mining and time.
I found that the you have to use the usb risers that are powered in the back of the usb riser board for the 12x pcie slots on the motherboard instead of the flex cables. I love the flex cables, but they often only work on the 1x pcie slots. See if this works for you and let me know. Thanks !!! I will try that. I have USB riser, but I was using the flex cable for the 12x 16x slot. Appreciated info EDIT: what is very strange is whatever card is at position GPU 1, it always the one who cannot create the DAG (Open CL error 36) or just doesn't hash (0 MH) or just hang after 1-2 hours. I got trouble with GPU2 and GPU5, but I can see that it's not the same card at the same position all the time since cards crashes and window re-install the driver automaticly or whatever he do it seems to shuffle the cards. If I run 4 cards, it is just more stable (2 days). I tried 2 MOBO (H81 PRO BTC) with the same result pretty much. I managed to get all 6 cars running 2 times, 10 mins and 2 hours with with only bootstrap Bios updates. I don't play with wattman, it is a waste of time at the moment, everything reset all the time. I red some info in another thread ... I will try what he said ... plug one card, reset CMOS, plug the rest (and some driver manipulation). I might just re-install Win 10 first since i have reinstalled 9.2 and 10.1 like freaking 20 times. airsick mobs are nutty. I just got this mobo http://www.newegg.com/Product/Product.aspx?Item=N82E16813157681 99 usd 30 rebate = 69 usd. I got it to replace the mob flooded at the solar array. So I wanted to run windows 7 and I got the usb mouse won't work bug. it works once you load all the drivers on the mobo cd/dvd so I go look around the house and find both a keyboard and a mouse with p2 jacks not usb jacks just to get to load the mobo drivers. spent an extra hour last night looking for the ps2 gear. the board is now working with 2 msi rx 470s' this was a cheap build. 2 cards on sale 370 usd. the mobo on sale 69 usd. so cost was 439. all other parts were salvaged from the array mobo. the ram the ssd the cpu. So I am up to just under 400 mh. I am happy I have chosen to be high on btc vs eth or anything else. I have just been converting all eth to btc at the moment. I also sold off a bit of btc to shore up my cash share of the game. I am kind of busy helping my neighbor fix up his garage. we tore out all old insulation put in new. then put in decent tongue in groove pine for the walls. next we will put up shelves. We should finish the job on this Friday. I'm death frustrated at the moment ... I was switching to USB riser one by one ... At the 4th card. Boom smoke. No idea why ... Card not recognized anymore. At least the MOBO is fine ... Cannot use Newegg replacement ... out of 30 days. 30 days I'm struggling with this shit. Opnened an RMA with Gigabytes ... cross fingers ... At least ... i'm now hashing with 5 cards ....
|
|
|
That's an interesting summary. I'm sure you are right in that the board temps on the optimizer would be lower than the board temps on the J4bberwock 4K board under load due to the extra copper. I just haven't had experience or reports of even a single yellowed plug on the 4K boards even under heavy loads (if anyone has one please post up!), and only occasionally on high-draw high-ambient 2880W boards as yours did with your modified cooling fans in a USPS shipping box and mine running single cables to Neptune cubes (which were prone to starting fires for their inadequate design). I'm going to see if I can melt a 4K board with a single DPS-2500 loaded as high as I can before hitting thermal shut-down to see if there really is a safe limit on them, because my experience is there is no safety concern with them whatsoever, and the hundreds (or thousands?) of customers with 4K boards in their hands from myself and J4bberwock would confirm that. Your report that it felt " a bit warmer to the touch." in this worst-case-scenario would seem to agree with that as well, and yet you still seem inclined to push the Optimizer boards for "safety". To each their own, it is a free market, but I feel that you are misleading the mining population with your assessment. Do you have a sales, commission or referral deal with Optimizer by chance?
I have one, replaced about 5 cables (Only the connector board side was over-heating, nothing miner side). I have now a constant airflow on them, just to be sure. My room temperature was high and a load of about 80%. I think the room temperature during the summer did not help.
|
|
|
I will re-install win 10 one last time and try the 6 card one rig ... if it fails it'll be 3 GPU per MOBO.
Tired of loosing hours of mining and time.
I found that the you have to use the usb risers that are powered in the back of the usb riser board for the 12x pcie slots on the motherboard instead of the flex cables. I love the flex cables, but they often only work on the 1x pcie slots. See if this works for you and let me know. Thanks !!! I will try that. I have USB riser, but I was using the flex cable for the 12x 16x slot. Appreciated info EDIT: what is very strange is whatever card is at position GPU 1, it always the one who cannot create the DAG (Open CL error 36) or just doesn't hash (0 MH) or just hang after 1-2 hours. I got trouble with GPU2 and GPU5, but I can see that it's not the same card at the same position all the time since cards crashes and window re-install the driver automaticly or whatever he do it seems to shuffle the cards. If I run 4 cards, it is just more stable (2 days). I tried 2 MOBO (H81 PRO BTC) with the same result pretty much. I managed to get all 6 cars running 2 times, 10 mins and 2 hours with with only bootstrap Bios updates. I don't play with wattman, it is a waste of time at the moment, everything reset all the time. I red some info in another thread ... I will try what he said ... plug one card, reset CMOS, plug the rest (and some driver manipulation). I might just re-install Win 10 first since i have reinstalled 9.2 and 10.1 like freaking 20 times.
|
|
|
I will re-install win 10 one last time and try the 6 card one rig ... if it fails it'll be 3 GPU per MOBO.
Tired of loosing hours of mining and time.
|
|
|
Holy Fudge,
Got my 2 replacement (one today and one yesturday).
Plug my new MSI yesturday ... tweak the Bios ... Start mining and it crashed after 1 hour ... restart the process did some tweak and crashes after 8 hours. I managed to get 40 hours uptime with 4 cards.
Now... I lost the control again. I'm doing clean driver un-install and re-install and i'm getting Code 43 on a card now.
I start mining process and I can see with memory the process use if it's fine or not. If i don't get to 300megs it's a no go ... Most of time now, the process just hang and that card got a Code 43 when I reboot.
This is impressive, hard to believe it's that hard to run 6 freaking GPU without troubles.
I got my Asus replacement, I'm starting to believe is it not a GPU issue but my MOBO.
I will do futher testing. I got another H81 PRO BTC. I will try it if it doesn't work anymore.
EDIT: I think i'm gonna have to re-install win10 again, I cannot remove the driver even with the cleanup exe from AMD. Win 10 re-install the latest driver after the cleanup and even if I manually remove the device and select delete the driver. This is quite Windows ;-)
Tried Motherboard B, had troubles ... switched back the PCI-e option in bios from GEN 1 to Auto. Boom 6 cards... Let's hope it's not luck !
|
|
|
Holy Fudge,
Got my 2 replacement (one today and one yesturday).
Plug my new MSI yesturday ... tweak the Bios ... Start mining and it crashed after 1 hour ... restart the process did some tweak and crashes after 8 hours. I managed to get 40 hours uptime with 4 cards.
Now... I lost the control again. I'm doing clean driver un-install and re-install and i'm getting Code 43 on a card now.
I start mining process and I can see with memory the process use if it's fine or not. If i don't get to 300megs it's a no go ... Most of time now, the process just hang and that card got a Code 43 when I reboot.
This is impressive, hard to believe it's that hard to run 6 freaking GPU without troubles.
I got my Asus replacement, I'm starting to believe is it not a GPU issue but my MOBO.
I will do futher testing. I got another H81 PRO BTC. I will try it if it doesn't work anymore.
EDIT: I think i'm gonna have to re-install win10 again, I cannot remove the driver even with the cleanup exe from AMD. Win 10 re-install the latest driver after the cleanup and even if I manually remove the device and select delete the driver. This is quite Windows ;-)
|
|
|
Hello; I have installed 6 graphic cards on Windows 7 x64. Four graphic cards work well. But the other two cards dont work. I use USB 3.0 risers with power. I couldnt figure the problem out. Does anyone has any information about it ?
Win 7 support only 4 GPU be default ... you need to "hack" the registry to able to run 6 cards. From my personal experience, with the hack I only managed to get 5 card running, the last one was not working with the Code 43 (probably the same thing you got) If you want to resolve quickly your situation, I propose that you install Win 10. I think you can run everything even if it's not activated with a Key.
|
|
|
wish I had a dollar for every time I skipped a word or two. I would be rich !
|
|
|
Or, since it's taking a billion hours to get all the sheathed cables assembled (have I mentioned before that sheathing sucks? Seriously doubles or triples total assembly time) I'll get orders shipped on Monday. Gotta leave for errands and burgers in an hour and that's not enough time to finish cables, let alone finish board assembly.
No stress for mines, you can ship me last !
|
|
|
i will apreciate if you can share more detail to me in Private message for trying your setting, and post the result. of course i will not share the room. i have some vga, powercolor rx 470 4gb , saphire nitro rx 470 4gb and 8gb, and rx 480 refference.
just asking, your mod is running smoothly in all card ? because each card have diffirent assic stat's
Please communicate as much as possible here so others can learn. I will swap ROMS etc by PM. Most of my ROMs are meant for mass use at my farm and run stable. I would recommend letting me get a couple together like the Reference ROMs and let you try it out and see what you think. Give me a day to put that one together. The ROMs I do I typically keep to myself but I am angry about the way people are treated on this forum by others. /Highfive
|
|
|
Just a heads up, further testing pointed to the Asus card and not the Power Color.
That card cost me another one since i had to do so much manipulation. I was dead tired yesterday and did a stupid mistake.
Hope the RMAs goes well.
I lost so much time with that Asus ...
Sorry to read this. Yeah I never like asus or gigabyte gpus. Powercolor Msi Sapphire Xfx Pretty good luck Asus's RMA not yet approved but the other card has been approved. Can't wait to get the replacement ! EDIT: Couldn't get a MSI for Asus as it is not refundable ... gonna have to life with it... I red somewhere here that hey use a different chip to control power on the card. Someone managed to get around 26 MH with it ... it might just order an MSI and sell the new card locally. EDIT2 Does anyone using Rx470 strix cards? I need some help to get lower power comsuption on my rig. I have 5 Asus Strix rx470 4gb cards and when I tested every single card, every card worked fine with settings: 1100/1850 and -40mV (26 mhs). But when I put all 5 gpus in rig and start mining cards demand valtage at least + 54mV. Gpus temperatures are also high because of this problem. I using 1500 memory straps and MSI AB for overclocking. If anyone have Idea how to decrease power comsuption on my Asus gpus please let me know.
The Asus cards use a custom VRM chip, which I've posted about earlier in this thread. The only work-around I've found is reducing VDDCI along with VDDC. I'm using the 1425Mhz strap at 1030/1875 and get a bit over 26Mh/s. EDIT: I still believe my card was defect ... throwing invalid share at stock freq, stock Bios it is not normal.
|
|
|
PCBs and all components have arrived. I spent today verifying everything and setting up the pick-and-place for bulk manufacture. I ran one off to test calibrations, completed assembly on it and it's running right now with about 10A of output current on 5V with no trouble.
Tomorrow I'll do all the surface-mount for the batch and start on through-hole.
I got started cutting wires to run out cables last week but ran out of wire; my reorder has arrived. So tomorrow I'll also be finishing up cutting wire and start cable manufacture. I've got some better heatshrink than I had available for the test cables MarkAz got, which means the production batch should come out better. As of right now there doesn't look to be any reason I can't get initial orders shipped out Friday.
I can wait to get my hands on them ! Thanks
|
|
|
I bought a few more of these RX cards and since I exceed more than 4 GPUs in Windows 7, I had to use that 7GPU Mod Registry hack, however it doesn't seem to work with the newest drivers 16.9.2
It worked perfectly fine running 6 - R9 280X on version 15.7.
The cards are recognized in Windows Drivers however, I can't get more than 5 GPUs to run in one system. Whats everyone been doing?
Install Win 10. I was stuck at 5 cards, last one was Code 43 EDIT: with 16.9.2 and the last one 16.10.x
|
|
|
Just a heads up, further testing pointed to the Asus card and not the Power Color.
That card cost me another one since i had to do so much manipulation. I was dead tired yesterday and did a stupid mistake.
Hope the RMAs goes well.
I lost so much time with that Asus ...
|
|
|
Anyone else has a Powercolor RX 470 ?
Mine is a piece of ****, crashing my rig all the time. I'm not even able to overclock it more the 22 MH.
I'm at job and i can't reboot my PC at the moment ... rig will be down for like 6 hours. Pissing me off
I'm going to ask for a replacement.
EDIT: RMA open, this is will go back home
I think the best card now are Sapphires and Asus, in terms of quality of components and field feedback. For quite sometime, I have avoided Powercolor and XFX ... thanks for this, i will keep that in mind for the future. So far my XFX is running fine
|
|
|
Anyone else has a Powercolor RX 470 ?
Mine is a piece of ****, crashing my rig all the time. I'm not even able to overclock it more the 22 MH.
I'm at job and i can't reboot my PC at the moment ... rig will be down for like 6 hours. Pissing me off
I'm going to ask for a replacement.
EDIT: RMA open, this is will go back home
|
|
|
May I point out this: We are at -3 % for the next jump and coins opened Oct at 612 usd they are now 639 usd So with the breakeven luck of 100% and the numbers above Oct 2016 has been excellent month so far. Good old time, feels like the Diff thread Good point Phil ! If we can keep the next one negative, it'ld be great !
|
|
|
So with 150 mh, I should get 6 more cards to mine at ETHPool ... LOL !!! Thanks to all of you, my rig has been up for 18 hours, rock solid. EDIT: Anyone tried to get a replaced with a card with a "bad ASIC" ? 69.1% Anyone has a lower ASIC then that ? EDIT2: I'm 4 days away from my first block EDIT3: 150.0 MH/s | 164.0 MH/s | 140.0 MH/s (Getting there !)
|
|
|
If you are referring to WattMan, it's built into the AMD driver: Gaming -> Global Settings -> Global WattMan.
It's really great except that it reset the settings upon any crash I have. If I could save/load profiles, it'll would be a nice workaround.
|
|
|
Phil,
The guy from Amazon just sent me a replacement board to test even I didn't ask for it. (I said I might have an issue with his MOBO but to wait a few more days for me to test and play with it.
and he sent 6 free risers ... that guy is EPIC ! (USB)
Herb
|
|
|
|