CryptoCurrency Mining with GPUs *spawn*

Seems coins are tanking a bit. Difficulty going up, prices going down. Probably better to trade coins than actually mine them right now, unless you already have a large rig.

Good time to buy into the market and diversify with some quality alt-coins like aelf or ecc.
I'm betting both BTC and ETH surge within the next 60 days. I won't be at all surprised if there is a big "correction" around late summer though.
 
Samsung enters crypto-currency chips business
Samsung Electronics has revealed it is making chips designed specifically to harvest crypto-currency coins.
The firm made the disclosure in its latest earnings report, where it said the activity should boost its profits.
...
For now, Samsung is providing little detail about its new crypto-currency business.
"Samsung's foundry business is currently engaged in the manufacturing of crypto-currency mining chips," it said in a statement given to the BBC.
http://www.bbc.com/news/technology-42892380
 
Samsung enters crypto-currency chips business
http://www.bbc.com/news/technology-42892380

Make no mistake, they are only a chip foundry here, not a developer.

There are three levels of cooperation:
1. Foundries like Samsung, TSMC and GlobalFoundires;
2. Solution providers like RamBus, eSilicon, Open Silicon, Global Unichip etc. who license HBM/HBM2 memory controller IP for a choice of foundries and process technologies;
3. End clients who actually design and order production of cloud computing ASICs.

https://www.esilicon.com/company/news-events/press-releases/esilicon-tapes-deep-learning-asic/
https://news.samsung.com/us/samsung...network-processor-rambus-28g-serdes-solution/
https://www.rambus.com/blogs/globalfoundries-demonstrates-2-5d-high-bandwidth-memory-hbm-solution/
https://www.globalfoundries.com/new...ates-25d-high-bandwidth-memory-solution-data/
http://www.open-silicon.com/open-si...ive-ip-subsystem-solution-for-high-end-netwo/
http://www.globalunichip.com/en-global/news/pressDetail/HBM2_Total_Solution/


This Korean site assumes Samsung is making these mining ASICs for some undisclosed Chinese company.
http://www.thebell.co.kr/free/content/ArticleView.asp?key=201801290100056740003492
 
Hopefully this at least puts to rest the "AMD is allocationg dies for Ryzens/TRs instead of GPUs" sentiment that I saw going around here.
Not really expecting inventory for gamers / hobby miners to appear anytime soon
Why would it put that idea to rest? Ryzen demand has likely dropped a bit from peaks at launch and unless ram production has spiked since all GPUs went out of stock, there is little reason to only now increase production. With record sales from Vega and Polaris alongside the Zen launch they may have been constrained previously. APUs in that picture as well and they could have been fulfilling other contracts prior to discrete.
 
Why wouldn't it? , now that AMD explicitly said that they could have manufactured more GPUs if they wouldn't have had memory supply problems.

They weren't willing to negotiate with memory makers to provide larger orders before, but they're willing to do it now.

There's plenty of GDDR5 and HBM2 available to whomever pays accordingly. AMD decided they wouldn't stand to gain a lot from entering a bidding war for the past 6 months, but apparently that's about to change.
 
Anyone understand what's going on with ETH? The difficulty is nearly back to the level at the end of the bomb last year. Is this something that is planned to continue, or a result of more miners? I always thought that the load balancing for the number of miners was separate to the difficulty?
 
What is this load balancing you are talking about?

See this formula :

https://ethereum.stackexchange.com/a/1910
So parent_diff, block_timestamp & parent_timestamp seem to be relevant. The point is that the block time for the network is supposed to be, on avereage, 14.75 seconds . So , each block readjusts the dificulty so to correct if the current block was computed too early or too late ( block_timestamp - parent_timestamp ) .


While I've not tracked the exact difficulty numbers, I can't say I've noticed anything unusual like you seem to have
 
Why wouldn't it? , now that AMD explicitly said that they could have manufactured more GPUs if they wouldn't have had memory supply problems.
Where did AMD say they weren't capacity limited most of last year? I only saw reference to the last quarter where production normally slumps following back to school and holiday season.
 
I really wonder how a lot of people say that they make money with mining right now. The difficulty is so high on monero and eth, and the prices are not so good. If you're still using an "old" mining rig already paid I get it, but buying new card right now seems crazy to me... A vega took about a month for 0.3xmr. And it's good at it...
 
Background: 2x miners each with 6 GPUs, one RX580 and the other GTX1070. They've been making $600-1250 USD/month since I built them. Went into the black about six months ago and I have relatively cheap electricity at 10.5 cents/kWhr. Today their estimated monthly earning is down to $800 after being at an all time high a week or two ago. ETH difficulty has definitely increased with the latest round of craziness (referring to people building mining rigs with $1000 GPUs as I don't see them ever making a net profit). For a while I mined ZCL and traded it for BTC or ETH with the 1070s but the 580s are still consistently best with ETH.
 
For a while I mined ZCL and traded it for BTC or ETH ...
Oh noes .. ZCL is going to grow.. at least till it becomes BCP.

I was stupid enoungh to trade away 24 zcl when they were worth 50$-70$ in total. Then the fork hadn't been planned and the coin just seemed dead . EZ $2000 lost
 
What is this load balancing you are talking about?

See this formula :

https://ethereum.stackexchange.com/a/1910
So parent_diff, block_timestamp & parent_timestamp seem to be relevant. The point is that the block time for the network is supposed to be, on avereage, 14.75 seconds . So , each block readjusts the dificulty so to correct if the current block was computed too early or too late ( block_timestamp - parent_timestamp ) .


While I've not tracked the exact difficulty numbers, I can't say I've noticed anything unusual like you seem to have

I was looking at the Coinwarz ETH difficulty chart; https://www.coinwarz.com/difficulty-charts/ethereum-difficulty-chart.

I'm on Nanopool and have a pair of 1070's mining ETH at just over 60MH/s average, most of the time. My payout is set to 0.2 ETH, and while at the end of last year I was getting paid every 19/ 20 days, now it's looking like i'll be over 30 days till my next payout.

To be honest, I'm not too sure where I got my ideas about load balancing from, probably somewhere on REDIT lol. But I thought the difficulty was set by an algorithm, like how they programmed in the diffuclty bomb ready for the move to POS that didn't happen, and then on top of that, the network adjusted the complexity of the mining based on the hashrate available to keep the blockchain time within certain parameters?

I must admit I am confused who is buying all these massively overpriced VGA cards where there doesn't seem to be much profit in mining even if your rig is payed for. I think I calculated that a Vega64 on ETH (I understand at the moment there are other choices for coins) would take about 7 months to pay for itself, even if you already had the suitable rig to stick it in. And that was assuming the difficulty doesn't increase. That is one hell of a gamble, even more so if you are buying a couple of full rigs.
 
I was looking at the Coinwarz ETH difficulty chart; https://www.coinwarz.com/difficulty-charts/ethereum-difficulty-chart.

I'm on Nanopool and have a pair of 1070's mining ETH at just over 60MH/s average, most of the time. My payout is set to 0.2 ETH, and while at the end of last year I was getting paid every 19/ 20 days, now it's looking like i'll be over 30 days till my next payout.

Well yeah, nothing unexpected . Difficulty did raise as the hashrate increased. And the payouts themselves devaluated further as the coin itself also decreased in value


To be honest, I'm not too sure where I got my ideas about load balancing from, probably somewhere on REDIT lol. But I thought the difficulty was set by an algorithm, like how they programmed in the diffuclty bomb ready for the move to POS that didn't happen, and then on top of that, the network adjusted the complexity of the mining based on the hashrate available to keep the blockchain time within certain parameters?

Well that's the algo (in the link) which sets the difficulty from block to block. They additionally did corrections at fixed given blocks as you said, the difficulty bomb. So it's not impossible well see some more of that in the future, althought I don't think anything is further planned

I must admit I am confused who is buying all these massively overpriced VGA cards where there doesn't seem to be much profit in mining even if your rig is payed for. I think I calculated that a Vega64 on ETH (I understand at the moment there are other choices for coins) would take about 7 months to pay for itself, even if you already had the suitable rig to stick it in. And that was assuming the difficulty doesn't increase. That is one hell of a gamble, even more so if you are buying a couple of full rigs.

Well there is some profit , at least for us which have electricity cheaper than 0.15$ / kwh. Tommorow it could be way more. Or it may be even a loss. This is crypto.
Also, as I keep saying on these forums, you don't have to sell your coins instantly. So what if the profitability is low, I hold the coins until they reach a price I'm confortable to sell at. For example if ETH comes goes to 1500-2000 $ (which it will most definetly will, this year) and we held on to it.. we'll all be laughing at these profitability numbers shown by sites like whattomine
 
I really wonder how a lot of people say that they make money with mining right now. The difficulty is so high on monero and eth, and the prices are not so good. If you're still using an "old" mining rig already paid I get it, but buying new card right now seems crazy to me... A vega took about a month for 0.3xmr. And it's good at it...

Nvidia profitability is still OK because it can mine Equihash coins quite effectively (ZCash, ZClassic, Bitcoin Gold). They tend to take turns at the top of the profitability charts as the hash rates move between them. Not sure what's up with AMD, since I'm not mining with any of those. I do know that a miner with a Lyra2Re2 kernel written directly in GCN assembly is in development for Vertcoin (that will probably be able to mine Monacoin and Verge as well) that, depending on how the hash rates turn out, may make those coins a viable alternative for AMD miners.
 
Back
Top