@ OP Please start using the code tag. It looks like this # and make log text walls into a much more readable scrollbox format like this. Oct 04 00:44:13 rockchip cgminer[966]: set_vol_on_i2c_b1p: set chain: 1, vol: ff,
Oct 04 00:44:17 rockchip cgminer[966]: B1 create dm_fanctrl_thread
Oct 04 00:44:17 rockchip cgminer[966]: B1: checking B1 chain
Oct 04 00:44:17 rockchip cgminer[966]: chain0: power down
Oct 04 00:44:19 rockchip cgminer[966]: fan0 speed:6030, fan1 speed:0
Oct 04 00:44:19 rockchip cgminer[966]: Fan speed too low: Fan0: 6030, Fan1: 0, ratio0:67, ratio1:0, (0x5a, 0x1, 0x1e, 0x4b, 0x3, 0x0, 0x0)!!!
Oct 04 00:44:19 rockchip cgminer[966]: chain0: encore first mode
Oct 04 00:44:19 rockchip cgminer[966]: Fan speed abnormal: Fan0: 6030, Fan1: 0, ratio0:67, ratio1:0, (0x5a, 0x1, 0x1e, 0x4b, 0x3, 0x0, 0x0)!!!
Oct 04 00:44:22 rockchip cgminer[966]: 0: detected 72 chips
Oct 04 00:44:22 rockchip cgminer[966]: collect core success
Oct 04 00:44:22 rockchip cgminer[966]: chain0: pre init OK, retries:0
Oct 04 00:44:22 rockchip cgminer[966]: chain1: power down
Oct 04 00:44:24 rockchip cgminer[966]: fan0 speed:6030, fan1 speed:0
Oct 04 00:44:24 rockchip cgminer[966]: Fan speed too low: Fan0: 6030, Fan1: 0, ratio0:67, ratio1:0, (0x5a, 0x1, 0x1e, 0x4b, 0x3, 0x0, 0x0)!!!
Oct 04 00:44:24 rockchip cgminer[966]: Fan speed abnormal: Fan0: 6030, Fan1: 0, ratio0:67, ratio1:0, (0x5a, 0x1, 0x1e, 0x4b, 0x3, 0x0, 0x0)!!!
Oct 04 00:44:24 rockchip cgminer[966]: chain1: encore first mode
Oct 04 00:44:27 rockchip cgminer[966]: 1: detected 72 chips
Oct 04 00:44:27 rockchip cgminer[966]: collect core success
Oct 04 00:44:27 rockchip cgminer[966]: chain1: pre init OK, retries:0
Oct 04 00:44:27 rockchip cgminer[966]: chain2: power down
Oct 04 00:44:29 rockchip cgminer[966]: fan0 speed:5970, fan1 speed:0
Oct 04 00:44:29 rockchip cgminer[966]: Fan speed too low: Fan0: 5970, Fan1: 0, ratio0:66, ratio1:0, (0x5a, 0x1, 0x1e, 0x4b, 0x3, 0x0, 0x0)!!!
Oct 04 00:44:29 rockchip cgminer[966]: Fan speed abnormal: Fan0: 5970, Fan1: 0, ratio0:66, ratio1:0, (0x5a, 0x1, 0x1e, 0x4b, 0x3, 0x0, 0x0)!!!
Oct 04 00:44:29 rockchip cgminer[966]: chain2: encore first mode
Oct 04 00:44:32 rockchip cgminer[966]: 2: detected 72 chips
Oct 04 00:44:32 rockchip cgminer[966]: collect core success
Oct 04 00:44:32 rockchip cgminer[966]: chain2: pre init OK, retries:0
Oct 04 00:44:32 rockchip cgminer[966]: chain3: power down
Oct 04 00:44:34 rockchip cgminer[966]: fan0 speed:6030, fan1 speed:0
Oct 04 00:44:34 rockchip cgminer[966]: Fan speed too low: Fan0: 6030, Fan1: 0, ratio0:67, ratio1:0, (0x5a, 0x1, 0x1e, 0x4b, 0x3, 0x0, 0x0)!!!
Oct 04 00:44:34 rockchip cgminer[966]: Fan speed abnormal: Fan0: 6030, Fan1: 0, ratio0:67, ratio1:0, (0x5a, 0x1, 0x1e, 0x4b, 0x3, 0x0, 0x0)!!!
Oct 04 00:44:34 rockchip cgminer[966]: chain3: encore first mode
Oct 04 00:44:37 rockchip cgminer[966]: 3: detected 72 chips
Oct 04 00:44:37 rockchip cgminer[966]: collect core success
Oct 04 00:44:37 rockchip cgminer[966]: chain3: pre init OK, retries:0
Oct 04 00:44:37 rockchip cgminer[966]: platform re-init: chain_num(4), chip_num(72)
Oct 04 00:44:37 rockchip cgminer[966]: sys : platform[5] miner_type[7] chain_num[4] chip_num[72] Cheers, |
|
|
|
|
Does your pool report the same hash rate?
Having the miners - from anyone - report the wrong speed is not unheard of. I had a s9 that reported 20+THs and have an A10 (33T) that insists it's running at 90T. Go fig. Don't think I've ever had one report too low but I can see it happening.
|
|
|
|
Read the 1st post in this thread... Specifically Quote from: -ck on November 16, 2017, 10:56:16 PM Mining BITCOIN is done exclusively with dedicated BITCOIN mining hardware based on ASICs - https://en.wikipedia.org/wiki/Application-specific_integrated_circuit . You CAN NOT meaningfully mine bitcoin today with CPU, GPU or even FPGAs. Bitcoin difficulty adapts to match the amount of mining done on the network and has reached levels trillions of times too high to mine meaningfully with PCs, laptops, tablets, phones, webpages, javascript, GPUs, and even generalised SHA hardware. You will not find software in this section to help you mine bitcoin in this absurdly inefficient manner in this subforum. It would cost you thousands of dollars in electricity per year to earn only a few cents in bitcoin. Even if you combined all the computers in the world, including all known supercomputer, you would not even approach 0.1% of the bitcoin hashrate today. Any discussion outside of ASIC related mining, except in the interests of academia, will be moved to the altcoin mining section. There isn't any point attempting to mine bitcoin with CPU or GPU even in the interests of learning as it shares almost nothing with how bitcoin is mined with ASICs and will not teach you anything Mining BTC with anything other than an ASIC-based miner is out of the question. For minable altcoins, well ask in the altcoin area of the Forum... |
|
|
|
Actually Frodocooper complained about these bespoke psu's back in 2018 when the s15's came out. And as I said back then: There are 2 reasons for it. 1. Using more PCIe connectors to feed a multi-kw mineris a bad idea. Too bulky and already proving needlessly unreliable because of the connectors being pushed to their limit and adding insult to injury them often being plugged/unplugged many times over the life of a miner and or PSU. PCIe connectors have a very small number of mating cycles before they begin seriously degrading. 2. Wall plug power efficiency. Today's best Platinum 12vdc PSU's are what - 95% maybe 98% efficient? Now that 12v was being dropped down via a final regulator on each hash board to whatever Vcore is being used. Typical string voltage was what, 7.5 to maybe 9vdc? At best the on-board regulators were *maybe* 85% efficient so... End result is that by losing the 2nd stage Vcore regulator losses using integrated PSU's that directly set the voltage results in more TH/kW plus by using bus bars it eliminates the high current connection problems. Oh, being able to use a higher string voltage lets ya use longer strings of chips so more TH/Current drawn through the string. The main tradeoff is that the PSU is setting the Vcore string voltage for ALL of the boards. Um, since we all know that the ideal voltage for any single board can vary a fair bit, that means the tuning process ends up with the best *compromise* voltage to reliably run the weakest board even though the other boards may perform better with a different voltage. Now as to *why* BM, Canaan, et al can't seem to have a more reliable PSU built is a mystery to me. It ain't rocket science to design a multi-kw PSU... Looking at the APW8 schematics posted at https://www.zeusbtc.com/manuals/Antminer-APW8-Power-Supply-Repair-Guide.asp it looks like they use I2C to talk to the PIC in the PSU. Yes the PSU also seems to have two distinct ranges for the DC output: They describe it as 15.9-16.3v and then give an upper range of 19.92-20.5v. Point is, assuming it *does* use I2C and you know what format the commands are sent to it are, it should be easy for some 3rd party to make a compatible PSU. Just have your firmware be able to identify what the hash boards are & need for voltage and talk to the PSU... |
|
|
|
|
Wouldn't finding a drop-in replacement PSU mainly be a matter of knowing the specs for it? Namely:
Output voltage range
Output current
Max power out rating
And here's the most important bit -- how the controller talks to the PSU to set the voltage output fed to the hash boards and if/how the controller knows what the PSU is doing. Is it simple 0-5v analog coming from a DA converter or analog output port on the SoC? Serial coms like I2C? PWM signal? etc.
|
|
|
|
NotFuzzyWarm and some others who know power could probably chip in with better specifics. My knowledge is largely academic.
No, you pretty much nailed it. The efficiency of any decent large power transformer ranges from at least 97% and for premium ones is over 99%. That leaves only the rate plan as a variable when it comes to incoming voltage & power. Oh, and of course the 1-time cost of the transformer if you have to pay for it  If you are the one supplying the transformer this is certainly a case for spending more to get a premium high-efficiency one vs the cheapest you can get. Considering that all of the power will be going through it, when pulling say 100kva through it every 1% difference is 1kva saved or lost. Using a 30 day month that = 720kWh/month so that cost adds up fast. |
|
|
|
Although not the cheapest out there from an energy cost perspective I am very happy with DelMarvaBlockchain. https://delmarva.io/They are in the US and have been nothing but honest, friendly and responsive since day 1. Tell Matt I sent you! Did a quick look at their site and do like most of what they show though their miner prices are way out of line with Bitmain's prices. The *most* part is the picture they have of a heavy power circuit breaker 'box' https://i0.wp.com/delmarva.io/wp-content/uploads/2022/08/IMG-2954-scaled.jpg?resize=1536%2C2048&ssl=1There is no way in hell that setup should have passed the electrical inspection... That plastic box does not stand a chance of surviving a short across any buses inside of it... Electrical distribution enclosures are designed and rated to contain and re-direct the huge explosive force that results from a major cross-phase fault. |
|
|
|
also right now mining BTC is the best advice I can give you, do not go near GPU mining.
Why? Are there no other altcoins to mining using GPUs just because ETH went POS? Yes there are still scads of shitcoins that can be mined using GPU's. Of course most are also worthless or damn close to it. Thing is -- this is the Bitcoin area where the sole topic of discussion is BTC. Not the altcoin mining area where all discussion of alts belongs. |
|
|
|
|
And then there the issue of eBang miners historically being utter crap with zero factory support...
|
|
|
|
Since a certain cloud mining site ad will be no doubt be deleted by the mods, the key line from that ad is I have the solution for you. By the way, you won't like this solution. Instead of looking for the best asis mining to buy, I would advise you to invest that money in a mining site BULLSHIT! >99.9% of cloud 'mining' sites are Ponzi scams and have nothing to do with 'investing'. Stay far far away from ANY of them!. Period. |
|
|
|
Without a doubt the Canaan Avalons are more reliable plus Canaan has stock of them in the US so there is no added Import duty to their price when you order directly from them. https://shop.canaan.ioThis is canaan official website I presume? Yes it is their miner sales channel. The main company site is https://canaan.io/ where you can also look into their AI products. |
|
|
|
A good thread to start with is here. The key takeaway from that thread is to take all 'advice' from so-called 'Experts' you come across on youtube, twitter, et al with a large dose of salt. Beyond that - spend a few days scanning through the various thread here in the Forum. Most often the titles will give a good hint to what they are about. |
|
|
|
Without a doubt the Canaan Avalons are more reliable plus Canaan has stock of them in the US so there is no added Import duty to their price when you order directly from them. https://shop.canaan.io |
|
|
|
The recycling depends on composition of the battery. Figure anywhere from $0.12 to $0.40 per kilowatt hour for large recycling (read tons and tons) for smaller amounts it's by the pound.
-Dave ... Keep in mind almost ALL of these companies hide crucial info when they sell you a battery, i had an offer for german paid battries that were sold as 100ah, when diving deep into the specs it turns out that it can only give you that much if you discharge it for 100 hours, so if you plan on using it at full load for 2 hours, it is no more a 100ah, more like 70ah. ... Battries are rather complicated and very expensive, i ended up buying a diesel generator, assumed it would die on me in 5 years, accounted for service and repairs and despite all that it was a lot cheaper. Ja. one MUST always refer to the battery manufacturer website and factory data sheets for the full picture. Actually that applies to buying any power electronic device. Even reputable manufacturer authorized distributors rarely give all the details (but will provide the links you want). Random resellers of things, well..... There is a 3rd/4th solution as well: Storing the power in large banks of supercapacitors which have a massively long lifetime but do self-discharge or the latest variant of them, hybrid supercapacitors that combine supercaps & Li-ion batteries into 1 device. In essence the caps handle whatever amount of load be it short high-current or long term output with no change in apparent capacity. btw: That article delves quite nicely into what a capacitor is and how they work  Of course both are very expensive but to a point it is balanced by them essentially being zero maint for 10-20+ years. One available provider of them is https://powerquality.eaton.com/emea/supercapacitor/default.asp Per that page, The Supercapacitor back-up power solution is available for all of the latest 3-phase Eaton UPS products starting from 8kW single units up to 7700 kW parallel UPS systems providing back-up time from seconds to minutes. No doubt using a lower% of full power rating gives more time. In other words long enough for the overall much cheaper (for hours/days/more of running) diesel genset(s) to kick in. Direct link to their supercap only UPS solution http://pqlit.eaton.com/ll_download_bylitcode.asp?doc_id=30241 |
|
|
|
|
To be honest, modular plug PSU's all have 1 major problem: They are pushing the maximum current rating through the contacts and generally have only 18ga wires between the PSU and PCIe plugs.
Ok, that's 2 major problems...
That said, since you are talking about running at reduced power it may not be an issue, just periodically check to see how warm the PSU connectors get. Yes, I've had a few rather spectacularly fry over the years from when I used them to power s3's & s5's...
Other thing to keep in mind is that particular PSU has 2 independent output rails. DO NOT connect power from the 2 different rails to 1 hash board! They will be shorted together and fight each other.
|
|
|
|
|
Again, please take the time to at least scan through the thread. All your answers are here...
That said - a standard USB port can supply only 0.7 to 1A so yes a powered hub that can supply at least 2A per port aka a 'charging hub' is REQUIRED to run at a decent speed and a made-for-mining hub is required to overclock as the sticks can easily pull 3 or more amps.
|
|
|
|
|
Good lord. HDD cache software is still being sold?
Um, it pretty much became a useless bit of kit in the late 1980's once the various OS's started to mature and began to handle disk access more intelligently...
|
|
|
|
|
Blockchain technology and Bitcoin are 2 related but different things....
Yes Bitcoin uses a blockchain *but* it is intended and designed to mainly be for monetary transactions - NOT general records keeping.
That said, there are many many real-world uses of blockchain technology that ARE used for records involving that involve tracking food production & distribution, various electronic device production, municipal lighting, long-haul trucking, etc. Publishing is just another variant of that. The key point here is that the blockchains used are tailored to a specific purpose and none of them have anything to do with BTC.
|
|
|
|
Didn't read all the sub pages, can anyone please give me a quick review and simple reply to following question/s:
- which USB miner is the currently latest and best mode with best hashrate and what's a cheap source for purchasing it?
- can I run that USB miner also on any Linux OS, eg my desktop Linux computer or Raspberry Pi 2 ?
- what is a realistic overclock and hash rate for 24/365 use for that particular USB miner ?
thanks to everyone in advance! .... Um, since this thread is about the Gekkoscience Compac-F guess what the answer is....  Where to buy it is last paragraph in the 1st post. |
|
|
|
|
Show Posts
