Search content
Sort by
Showing 20 of 28 results by Tom17
When I do -i 6 on my 7950's, it says there is not enough ram and it reduces it to 5! What SOL/s are you getting on your 7950s?
Well I have managed about 260 (-i 5) at 950/1450/1.05v, last claymore zec miner.OK, card spacing makes a huge difference, but I guess you all know that. Finally have my headless miner (vnc, ugh) running with 2 x 7950's. Time to let it run as a POC to get wival approval to get more efficient cards 
Thanks for the help!
Thanks for the help!
I like the sound of that! But alas, it seems the bandwagon has started up again and I jumped on at the wrong time (again). All the good deals on kijiji are being snapped up...
Still one card gets hot and ramps up the fan speed. I'll try putting it together with my risers so they have more space to cool. I imagined cramped spacing would take longer to show up in heat/fan speed than a couple of minutes....
Thanks for your post. It makes a lot of sense and has put my mind at ease, even though I wasted a day on this lol.
Thanks for your post. It makes a lot of sense and has put my mind at ease, even though I wasted a day on this lol.
This makes sense.. The info was probably in this thread, but 700+ pages 
So if I plan to get any newer cards, I can't really mix either. Maybe I should offload these 950's and get something newer... Sounds like it will be much simpler lol.
Still having problems even in Windows. When I put in a 7950 and a 280x (Almost the same, both Tahiti etc), then the 280x always goes straight to full fan speed, but it's fine when it's in on its own.
I just did some card shuffling so I now have 2 Sapphire 7950's in the same box and the 280x is in my son's machine... Let's see.
I'm gathering I may be better off sticking to Windows then while I have these older cards...
So if I plan to get any newer cards, I can't really mix either. Maybe I should offload these 950's and get something newer... Sounds like it will be much simpler lol.
Still having problems even in Windows. When I put in a 7950 and a 280x (Almost the same, both Tahiti etc), then the 280x always goes straight to full fan speed, but it's fine when it's in on its own.
I just did some card shuffling so I now have 2 Sapphire 7950's in the same box and the 280x is in my son's machine... Let's see.
I'm gathering I may be better off sticking to Windows then while I have these older cards...
For the 7950, you might be better off going with UBuntu 14.04 and using the fglrx 15.12 (they are in one of the Ubunty repositories with "-302" as part of the name) drivers.
I'm not sure if the "PRO" stuff supports older cards, though in theory it SHOULD work with anything GCN.
I'm not sure if the "PRO" stuff supports older cards, though in theory it SHOULD work with anything GCN.
I actually went down that path too since making that post. I did away with the USB OS disk and found a regular disk. Tried a fresh UB 14.04 which was a fail, then tried UB 14.04 and installing the catalyst drivers failed. But I downloaded them instead of looking in the repo... I should try again with the repo versions I guess. Would using UB 14.04.5, would that have been a problem?
This was my error:
Code:
$cat /usr/share/ati/fglrx-install.log
NOTE: If your system has logged the missing packages required for installation, install them in the order as per the log file to resolve package-dependency issu
es.
Supported adapter detected.
Check if system has the tools required for installation.
Uninstalling any previously installed drivers.
Creating symlink /var/lib/dkms/fglrx/15.302/source ->
/usr/src/fglrx-15.302
DKMS: add completed.
Kernel preparation unnecessary for this kernel. Skipping...
Building module:
cleaning build area....
cd /var/lib/dkms/fglrx/15.302/build; sh make.sh --nohints --uname_r=4.4.0-78-generic --norootcheck.....(bad exit status: 1)
[Error] Kernel Module : Failed to build fglrx-15.302 with DKMS
[Error] Kernel Module : Removing fglrx-15.302 from DKMS
------------------------------
Deleting module version: 15.302
completely from the DKMS tree.
------------------------------
NOTE: If your system has logged the missing packages required for installation, install them in the order as per the log file to resolve package-dependency issu
es.
Supported adapter detected.
Check if system has the tools required for installation.
Uninstalling any previously installed drivers.
Creating symlink /var/lib/dkms/fglrx/15.302/source ->
/usr/src/fglrx-15.302
DKMS: add completed.
Kernel preparation unnecessary for this kernel. Skipping...
Building module:
cleaning build area....
cd /var/lib/dkms/fglrx/15.302/build; sh make.sh --nohints --uname_r=4.4.0-78-generic --norootcheck.....(bad exit status: 1)
[Error] Kernel Module : Failed to build fglrx-15.302 with DKMS
[Error] Kernel Module : Removing fglrx-15.302 from DKMS
------------------------------
Deleting module version: 15.302
completely from the DKMS tree.
------------------------------
In the end I threw Win10 on it and it just worked... (Except that the second card went to full fan speed almost straight away, will look into that).
I guess I need to try UB 14.0.5 again and try installing from the repo...
I'm having some trouble here. I have a rig that I set up back in 2013/2014 and used to mine it.
Back then, it was like so:
Gigabyte 1155 mobo with cheap Celeron and 4GB ram.
Booting Ubuntu 12.04 on a USB drive (Surprisingly this worked well)
Some version of the FGLRX drivers
Multiple 7950 cards
So I dusted it off recently to mine ZEC, installed Claymore on it and it worked straight away, but only saw one GPU at a time. Linux could see and use multiple GPU.
I decided it was time to just rebuild the OS. So I got a new USB stick, installed Ubuntu 16.04 server following this guide: https://blockoperations.com/build-6-gpu-zcash-headless-mining-rig-ubuntu-16-04-using-claymore/
It's using the latest AMDGPU-PRO driver (17.10, the link did not show the 16.50 drivers any more).
Right now, I am having zero luck. I have a single Sapphire 7950 in there, and the hash rate always shows as 0. No overclocking, no PCIx risers, everything stock.
I tried using the older drivers (16.50) as I eventually found them, but still no luck. In fact it was worse as it now saw the cards as Hainan instead of "Tahiti".
I have since read that maybe I should be using Ubuntu 14.04. But I am not sure where to head right now. I have wasted a whole day on this so far.
Is there a currently recommended way to get this all working in Linux with a 7950 card?
Back then, it was like so:
Gigabyte 1155 mobo with cheap Celeron and 4GB ram.
Booting Ubuntu 12.04 on a USB drive (Surprisingly this worked well)
Some version of the FGLRX drivers
Multiple 7950 cards
So I dusted it off recently to mine ZEC, installed Claymore on it and it worked straight away, but only saw one GPU at a time. Linux could see and use multiple GPU.
I decided it was time to just rebuild the OS. So I got a new USB stick, installed Ubuntu 16.04 server following this guide: https://blockoperations.com/build-6-gpu-zcash-headless-mining-rig-ubuntu-16-04-using-claymore/
It's using the latest AMDGPU-PRO driver (17.10, the link did not show the 16.50 drivers any more).
Right now, I am having zero luck. I have a single Sapphire 7950 in there, and the hash rate always shows as 0. No overclocking, no PCIx risers, everything stock.
I tried using the older drivers (16.50) as I eventually found them, but still no luck. In fact it was worse as it now saw the cards as Hainan instead of "Tahiti".
I have since read that maybe I should be using Ubuntu 14.04. But I am not sure where to head right now. I have wasted a whole day on this so far.
Is there a currently recommended way to get this all working in Linux with a 7950 card?
Re: [ANN] microCoin MRC - alcurEX - Scrypt-Jane - nFactor - Now on MintPal
I was thinking about the N-refactor earlier today...
If the number of miners stays approximately the same, and the effective hashrate reduces due to the N-refactor, the diff should go DOWN, right?
If the number of miners stays approximately the same, and the effective hashrate reduces due to the N-refactor, the diff should go DOWN, right?
Re: [ANN] microCoin MRC - alcurEX - Scrypt-Jane - nFactor - Now on MintPal
Exponential growth is always slow and steady before the 'storm' 
http://en.wikipedia.org/wiki/Exponential_growth
http://en.wikipedia.org/wiki/Exponential_growth
Re: [ANN] microCoin MRC - alcurEX - Scrypt-Jane - nFactor - Now on MintPal
Yeah I like the ramp up & down too. It's one of the things that attracted me to this coin. And the ramp up should finish after the exchange is open. By my calcs, it should be about 21-22 days before we reach the top of the mountain. Exchange should be open before then...
Also, has anyone worked out or posted an N Factor schedule? If not, I could work it out if needed.
Also, has anyone worked out or posted an N Factor schedule? If not, I could work it out if needed.
Re: [ANN] microCoin MRC - alcurEX - Scrypt-Jane - nFactor - Now on MintPal
Guys...
./microCoind getinfo
...
"moneysupply" : 5102804861.70803547,
...
2700 blocks a day at ~100,000 coins per block (currently) = 270 million new coins a day. Yes, we are small fish with our millions or tens of millions
BTW, this will go up to over 400 million per day before it finally starts riding down the other side and finally settles on 2.7 million new coins a day.
./microCoind getinfo
...
"moneysupply" : 5102804861.70803547,
...
2700 blocks a day at ~100,000 coins per block (currently) = 270 million new coins a day. Yes, we are small fish with our millions or tens of millions
BTW, this will go up to over 400 million per day before it finally starts riding down the other side and finally settles on 2.7 million new coins a day.
Re: [ANN] microCoin MRC - alcurEX - Scrypt-Jane - nFactor - Now on poloniex.com -
Wow, people are buying already? Thats' pretty cool
So... https://www.mintpal.com/voting
We were on top, but now we are in 2nd again with a sizeable gap. I guess someone threw some BTC at the DigiByte vote...
2 h ago, the MintPal twitter feed said 'taking the winner in a few hours' so that could be any time right? Maybe we should throw a bit of BTC at it to get it on top? I could do 0.01 but that wouldn't be enough. Any others willing to help us stay on top of this thing? (Except, I was just told it's a scam...)
Tom...
So... https://www.mintpal.com/voting
We were on top, but now we are in 2nd again with a sizeable gap. I guess someone threw some BTC at the DigiByte vote...
2 h ago, the MintPal twitter feed said 'taking the winner in a few hours' so that could be any time right? Maybe we should throw a bit of BTC at it to get it on top? I could do 0.01 but that wouldn't be enough. Any others willing to help us stay on top of this thing? (Except, I was just told it's a scam...)
Tom...
Re: [ANN] microCoin MRC - alcurEX - Scrypt-Jane - nFactor - Now on poloniex.com -
It (the community) starting to move. Yeaahhhhh
For the IRC channel, once numbers start to accumulate, I will set up my bot. It does all the usual stuff such as diff, perday, worth, prices etc but it also has a faucet/giveaways and coming soon is a multi-coin lotto
Come on in! https://kiwiirc.com/client/irc.freenode.net/#microcoin
Tom...
For the IRC channel, once numbers start to accumulate, I will set up my bot. It does all the usual stuff such as diff, perday, worth, prices etc but it also has a faucet/giveaways and coming soon is a multi-coin lotto
Come on in! https://kiwiirc.com/client/irc.freenode.net/#microcoin
Tom...
Re: [ANN] microCoin MRC - Scrypt-Jane - nFactor - Now on poloniex.com!
Hey all, I just discovered this today and might be interested in mining/buying.
I noticed that there is no #microcoin channel on Freenode. Was this a deliberate decision or just an oversight? Maybe I have the wrong channel name
Also, is that crycurex exchange safe? I just sent some BTC there to buy some MRC and already having issues - support is looking into it supposedly.
Thanks,
Tom...
I noticed that there is no #microcoin channel on Freenode. Was this a deliberate decision or just an oversight? Maybe I have the wrong channel name
Also, is that crycurex exchange safe? I just sent some BTC there to buy some MRC and already having issues - support is looking into it supposedly.
Thanks,
Tom...
Re: [ANN][VTC] Vertcoin - Adaptive N-factor in Scrypt - No more ASICs
I just read a bunch of threads about them. Not a hoax, but seriously overpriced still... This post pretty much sums it up...
AND They are not keen selling them. With prices ppl suggest here. If you have Machine laying golden eggs while none else have why would you sell it. Also this is totally GEN 1 hardware and totally in line what first SHA Bitcoin ASIC:s were performing. The Chip is not super performer but what it does it does it 1/40th of the Electricity of GPU rig
PPL are blinded by current GH and TH of BTC ASIC. If and when they get mass sampling of this chip what would stop them putting 160 chips in a Blade taking 70-100W and producing massive 9.6MH/s thats 12 GPU rig right there in 1 single ~70W blade.
Price? dont dream about it. They will mine and overprice the chip to cover R&D costs and get healty profit before dumping them into the market. If you are first you are First.
-Edit-
http://www.middlecoin.com/reports/1M3jtksp1upR33SX1VzeCfH5x9fc6zKykR.html
Been following above middlecoin address and they have add around 100MH/s daily there. Thursday - Friday - Monday - Tuesday - Today i just saw 957MH/s accepted. It is past 1GH/s accepted hashrate tomorrow evening. That would be closer 2000 AMD GPU units in mining rigs and rough numbers would be what 500-700 kw/h power and heat at minimum. I'm no professional but i would say it's not easy to obtain that much power. Compared these chips would do it with 7-15kw/h which is nothing.
I've seen that on Middlecoin, but I'm failing to see how this is transformational. It would seem to me that the key in the near term (at least months) is equipment price/kh, not kh/kw. Granted, I have cheap power in the US, but at the rates alt coins are paying, the cost of power is almost exactly an entire order of magnitude less than the revenue that can be obtained by that power. Case in point: 0.01 BTC/day/(MH/s) is pretty easy to reliably obtain. For 1Mh/s, you're looking at about 300-350 W. 0.01 BTC = about $8.5 currently (and stands to rise). 350W = $1.68 even at a very liberal $0.20/KWhr.
My point with all of this is that hashing power is clearly the dominant metric for profit at this point. With that said, the prices listed in this thread for ASIC hardware are about two-fold WORSE than typical cost/hash represented through GPU mining.
What this is all leading me to is this - if power could (even liberally) be assumed to be soaking a quarter of your revenue from GPU mining, and these products are twice the cost of GPU mining, there is literally NO break-even. Even if they use no power at all, the best you could do with them is still to bring two thirds of the profit/investment of GPU mining.
It seems that these become a true alternative when they get to between the 1.0x and maybe 1.25x the price/hash power, and truly preferable when cheaper still. Similarly, it would take until profits/hash power are halved from their current levels to have the same effect. With all of the pump and dump coins and stupid speculators betting on them in the short term coming and going, I don't see that happening any time soon.
Am I wrong?
AND They are not keen selling them. With prices ppl suggest here. If you have Machine laying golden eggs while none else have why would you sell it. Also this is totally GEN 1 hardware and totally in line what first SHA Bitcoin ASIC:s were performing. The Chip is not super performer but what it does it does it 1/40th of the Electricity of GPU rig
PPL are blinded by current GH and TH of BTC ASIC. If and when they get mass sampling of this chip what would stop them putting 160 chips in a Blade taking 70-100W and producing massive 9.6MH/s thats 12 GPU rig right there in 1 single ~70W blade.
Price? dont dream about it. They will mine and overprice the chip to cover R&D costs and get healty profit before dumping them into the market. If you are first you are First.
-Edit-
http://www.middlecoin.com/reports/1M3jtksp1upR33SX1VzeCfH5x9fc6zKykR.html
Been following above middlecoin address and they have add around 100MH/s daily there. Thursday - Friday - Monday - Tuesday - Today i just saw 957MH/s accepted. It is past 1GH/s accepted hashrate tomorrow evening. That would be closer 2000 AMD GPU units in mining rigs and rough numbers would be what 500-700 kw/h power and heat at minimum. I'm no professional but i would say it's not easy to obtain that much power. Compared these chips would do it with 7-15kw/h which is nothing.
specs are here
https://github.com/gridseed/gc3355-doc/blob/master/GC3355_DataSheet.pdf
And holy shit, it claims to do 59.6 KH/s while using 0.44 W
This is 0.00738 W/KH/s versus 0.300 W/KH/s for a GPU -- a 40.7 fold increase in efficiency. This is about the same increase in efficiency as compared to BTC when ASICMINER first introduced its chips; I'm curious to see how they did it.
https://github.com/gridseed/gc3355-doc/blob/master/GC3355_DataSheet.pdf
And holy shit, it claims to do 59.6 KH/s while using 0.44 W
This is 0.00738 W/KH/s versus 0.300 W/KH/s for a GPU -- a 40.7 fold increase in efficiency. This is about the same increase in efficiency as compared to BTC when ASICMINER first introduced its chips; I'm curious to see how they did it.
I've seen that on Middlecoin, but I'm failing to see how this is transformational. It would seem to me that the key in the near term (at least months) is equipment price/kh, not kh/kw. Granted, I have cheap power in the US, but at the rates alt coins are paying, the cost of power is almost exactly an entire order of magnitude less than the revenue that can be obtained by that power. Case in point: 0.01 BTC/day/(MH/s) is pretty easy to reliably obtain. For 1Mh/s, you're looking at about 300-350 W. 0.01 BTC = about $8.5 currently (and stands to rise). 350W = $1.68 even at a very liberal $0.20/KWhr.
My point with all of this is that hashing power is clearly the dominant metric for profit at this point. With that said, the prices listed in this thread for ASIC hardware are about two-fold WORSE than typical cost/hash represented through GPU mining.
What this is all leading me to is this - if power could (even liberally) be assumed to be soaking a quarter of your revenue from GPU mining, and these products are twice the cost of GPU mining, there is literally NO break-even. Even if they use no power at all, the best you could do with them is still to bring two thirds of the profit/investment of GPU mining.
It seems that these become a true alternative when they get to between the 1.0x and maybe 1.25x the price/hash power, and truly preferable when cheaper still. Similarly, it would take until profits/hash power are halved from their current levels to have the same effect. With all of the pump and dump coins and stupid speculators betting on them in the short term coming and going, I don't see that happening any time soon.
Am I wrong?