I wouldn't say 'never' on the R4 but yes it is a real roll of the dice. I have 3 of them. 2 perfect and near silent, the 3rd lost the top board after 24hrs. Gotta send that to Colorado some time now that warmer weather is here...

Ya, the s7-LN especially with sidehack's UV/UC mod are exceptionally quiet. Also only pull around 750-950w depending on how you set them up, db level is commensurate with speed/power.

eep
The only possible thing to try would be to get a couple cans of electrical contact cleaner or circuit board cleaner and hose them down to *maybe* dislodge crud. Problem is, if the boards are very old and most likely covered in fine dust then flooding them COULD MAKE THINGS WORSE if they are not pristine when done. You could very well force dirt into places it wasn't before.

And no, no how-to's beyond 'Does the red LED power lights on the board turn on?' and perhaps swapping the data cables around just to make sure it is the board(s).

BitmainWarranty in CO pretty much prefers folks don't try to fix boards themselves. I can imagine what some look like after folks ham-handedly try...

Speaking of which, had a s9 batch-13 board fail yesterday so time to send it to them.

***  Foxminers LLC is a SCAM! If someone buys their miners without a well-known and trusted Forum User 1st getting one for doing a review and posting it here then well, you have been warned ***Since sorta walking down memory lane but at least keeping this SCAM thread at the top for the perps to fret over, what the heck. If it costs them losing even 1 sale it is worth it. There. It's on-topic again

2112 you hit on soooo many good points. If you catch my other posts in other threads ya may have noticed I take the 'Teach a person to fish' approach to answering questions. LEARN folks! Don't just raise yer hand to get one specific answer, learn the why and ta-dum! things start to connect..

Again my profession isn't CS, since 1977 it has been design/application of industrial lasers. To me computers are just tools. Und damn we have come a long way from paper tape NC controls with literally relay based ladder logic...

The how/why knowledge of semiconductors themselves well, as a kid in late 50's was building vacuum tube toys and literally grew up tinkering with discrete transistors ect,, led to enlisting in the Air Force (dinna want to be drafted so choosing a service best option) > long haul microwave, ended up in the 1839 E&I Group out of Keesler AFB installing the gear for com sites- I tell ya, a tropo site is something to behold. Also most often on a tiny island in the middle of nowhere... https://en.wikipedia.org/wiki/Tropospheric_scatter No large use of satellites back then. While in the military, got BS EE for the heck of it.

Out of the military -- moved up the spectrum into lasers and been there ever since. Only burnout was 2 years of 80hr weeks doing field service at my 1st company Photon Sources (now Lumonics). My instant 'In" with them was the multi-kw microwave power background and um, I had build a dye laser in Jr. high school. Another engineer there was tired of it as well so we started SLI http://www.synchronlaser.com/ and will be there until I die. When you love what you do it isn't work!

The laser biz is my link to wassup in the chip manufacturing world and a helluva lot of other industries. Pure how to make things. Comes with the turf: With bleeding-edge products intended to leapfrog how-things-are-done-today, the customer does not exactly know what they need for producing them, only what they want to produce. Guess who gets to figure out how to do it since no no one else has yet done it outside of a lab... 1st step - understanding to the nth detail what they make that requires - unique solutions -- vs off-the shelf so I can have a handle on how my systems may affect the desired result. For power chips it may only be an interposer/heat spreader but the dies care a great deal about their interface to it so understanding their construction is mandatory.

Damn this thread goes a'wandering...

Sigh. Look at the Bitmain specs maybe?.....
Nice round ball park is 1,500w per-s9. Will need 208-240 volts and with that voltage the amps per-miner =  1500/voltage.
Your turn.

And that ^^ being right to the main point of why no 'new' way to hash faster is going to be seen soon is why I deferred to 2112. Only so many ways to skin a cat in hardware when it serves one and only one purpose.

I am not involved in the architecture design and strictly follow the node-size tech eg physical construction of the gates vs implementing actual circuits to explains 'why's' as part of the voodoo I do. My part comes in making it possible to efficiently pull the heat from the dies, Specifically making the systems that micro-machine vias into ceramic interposer/heat spreaders the dies get attached to as part of their packaging to become 'chips'.

Sure you can shrink gate sizes to pack more gates in. CPU, phone baseband chips, network fabric switches and such have the luxury of a bigger physical die size footprint to accommodate the plethora of connections also resulting in lower power-density from more area to spread out the heat/attach heat sink or other thermal route to. Mining ASICS -- sure can fit in more cores but methinks they are already at the point of diminishing return for packing density vs power needed to feed the die. That is why these are so bloody hot. With the BM1387 chips reporting die temps of 95C to 110C as 'normal, God knows what the real junction temps are.

Which pools? Hard proof or at least quotable suspicious actions by them leading to this theory?

Cute: Looks like an old s1 without the RiPi on it
Don't know if you caught the edit I did: As far as 7nm goes, this is that last I have come across https://arstechnica.co.uk/gadgets/2015/07/ibm-unveils-industrys-first-7nm-chip-moving-beyond-silicon/ Do note that 'chip' is referring to an assemblage of functional test structures - not a usable logic chip. Again, is a couple years old but things have been very quiet in terms of news released by the Foundries.

Has several good links in it to other articles about wassusp

That is a lot harder question than you may think as there are several layers to it... For one, no one else has anything approaching what they claim. The world of crypto ASIC design is very very small and frankly no one does anything without the others working along the same lines and being rather vocal about it. Since when it comes to the Next Wonder Miner all we hear from real makers is crickets, well....

On the physical construction and layout of an ASIC I have to defer to a member here with the handle 2112 who is/was an instructor in chip design. You might want to PM them about it. A couple papers on 16/14nm tech http://www.techdesignforums.com/practice/guides/14nm-16nm-processes/ and from 2013 http://www.eetimes.com/document.asp?doc_id=1280773 Somewhere I have several papers from Mentor Graphics on their toolchains for those nodes...

As far as 7nm goes, this is that last I have come across https://arstechnica.co.uk/gadgets/2015/07/ibm-unveils-industrys-first-7nm-chip-moving-beyond-silicon/ Do note that 'chip' is referring to an assemblage of functional test structures - not a usable logic chip.

To me the core <yes, pun> issue is, how many cores will fit in a chip? Unlike CPU's, crypto ASICS are extremely simple beasts. Each has serial coms, a smattering of working memory, and a buttload of hard-wired SHA logic cores. Whereas a CPU contains many different circuits for several kinds of IO, along with cache and math, together with the actual few to handful of CPU cores. The latest Intel Xeon has what, 12 physical cores in it? As I recall, Bitmains BM1387 chip used in the R4, s9 and T9 have 250 cores in them, the s9 uses 189 of those chips. In a way GPU chips are similar (high core count) but rather like FPGA their operations can be changed via programming but that ability again leads to speed and power penalties.

Since Bitmain has not released a data sheet for that chip, here is the specs for their last 28nm chip used in the S7 miner https://shop.bitmain.com/files/download/BM1385_Datasheet_v2.0.pdf to poke through. Much of it should still apply to their current 16nm chip

That simplicity does have a down side: Power density. Miner design moved away from large monolithic chips because it is very difficult to power and cool a chip which size-wise *could* these days hold several thousand cores and dissipate >1,000 watts. The BFL Monarch, Hashfast Minion, and a few other failures come to mind...

On the software end, it becomes fuzzier. There 2 things come into play, Stratum which works with the pools to create work and the miner software itself. Stratum docs https://bitcointalk.org/index.php?topic=557866.0

For the miner software, talk to -ck since he wrote cgminer which is what almost all miners use. Considering even the latest miners can run on a single RasPi-3 front-end that rather says then and there that optimizing code (even more as it *is* rather mature) will have little impact.

As for the Samsung 10nm processor: Read into the link and the one from Intel. To claim the 1st-to-market moniker the Samsung is kind of a cheater: Yes the gates are around 10nm but the metalization (connections) are the same 22nm they use with their 14nm chips. Intel tends to shoot for doing the entire process smaller - not just the gates.

*Could* Bitmain be working with TSMC to make a 12nm mining ASIC? (With the same 22nm metal layers they use in their 16nm FinFET's). Sure. When it comes to boutique chips like mining ASICS Bitmain is the one company that certainly has resources to pay for it. But makes no economic sense to me for them to do it: Their BM1387 is king of the hill with no competitors and after the beating the entire industry took finally getting 16/14nm to the consumer market coupled with the still-erratic chip-to-chip performance headaches they (and Avalon, and BitFury) have to deal with there is just-no-point to do it at this time. Even if they did, there is still the problem of boutique chips being last in line for the Foundry production priorities. Just as is still the case with 16nm chips first come the folks who financed ALL of the research involved eg, Apple, AMD, Cisco, Broadcom, et al.

Or refer to here [General] Bitcoin Wallets - Which, what, why?
 for a list of all types, advantages/disadvantages...

That ^^ and more to the point, in the altcoin sections there are scads of people who run the systems needed for alt's so that is the best place to get your answers. Here, it is BTC rigs only.

Just noticed this https://bitcointalk.org/index.php?board=39.0 I never knew that the Forum had a Beginners area! Given that, the Forum registration process should point folks to the Sticky there https://bitcointalk.org/index.php?topic=1689727.0 when they finish registering. Might save a lot of wrong-section annoyances...  

When done by the OP (Original Poster) it is because the topic has been fully answered and discussed to death.
More often threads are locked by the Moderators and that is usually because it is seriously off-topic or in the wrong section.

Once a fan is used the orientation of the heatsink is a moot point unless it is one very tiny low-flow fan. Keeping fins vertical only matters with convection cooling.

I do hope you have read my earlier posts about the power layout of the Plugable 7-port hub...

No one is even close to shipping production sub-14nm chips. See edit. a big Hmm here...
Engineering samples of test circuits to begin characterizing what to expect from them -- maybe. After the almost-there for a couple years hopes the 16/14nm showed (and even though in now production are still producing horrible yields) most industry pundits still put any 10nm production from even Intel/IBM to be early next year. Even then they still follow that with - maybe.

Edit: Just did a search on Samsung and turned up https://arstechnica.com/gadgets/2017/02/samsungs-got-a-new-10nm-octa-core-chip-with-gigabit-lte-for-flagship-phones/
And from March, http://www.eetimes.com/document.asp?doc_id=1331504
https://www.xda-developers.com/intel-claims-next-chip-generation-ahead-samsung/

wtf? There has been NO mention about that kind of progress in the IEEE Spectrum feeds I get. Those cover beyond-bleeding-edge tech and last I read 10nm was still in test-mode so this bears looking into...

Difference is, powerball players are not having to pay an electric bill each month - we miners do.
I have been running over 100THs since the end of last year (currently ~237THs) and hit 1 block on KanoPool last Feb. So unless Lottery mining and you have free electric, solo is NOT worth it. Being in a great pool like Kano on the other hand -- priceless.

The miner efficiency is irrelevant and only compares hash rate to power used, has nothing to do with heat load.

For any electric power heat source they are all 100% efficient. Power in = power out as heat.
Cooling is of course another matter as it involves work moving heat from one location to another, not creating heat. In that case efficiency does matter but only in the sense that it takes x amount of power turned into y amount of mechanical energy (with losses turned into heat) to move z amount of heat energy.

The miner GUI shows what??
Do all 3 cards show up? Are all chips showing as '0' and no 'x'?
As the s7's age they start dropping a card for time to time. If the s7 had been modified to use lower power it happens more.
Just cycle the power on/off and should come right back up to speed.

1st thought should be to Google 'convert watts to BTU'. Why does no one seem to know how to use search engines....
FWIW, BTU/hr=Watts x 3.4129

The CFM, fair question. Don't think there is any data on that. I believe the fans move 250CFM free-air but under load, pure guesswork. I'd ballpark it as around 150CFM

Hate to pop your shiny glowing bubble vision of What Will Soon Be...  No, we won't and I for one will be damned before I support mention that something along those lines is possible today or even within 2 years. All it does is lend credence to scammer claims. They rely on that bright wide-eyed wonder to pitch their frauds.

Even back in 2013 when 65nm miners were finally moving out the door and 28nm tech miners were being hawked/pre-sold, the current 14/16nm tech had been in development for around 5 years and they had only recently succeeded in making viable test circuits of any sort. It took another 3 years before TSMC, Samsung, and GloFo finally started commercial production of 16/14nm chips.

The next generation of chips targeting the 10nm and 7nm nodes are in the same boat. Just because Bitmain et al say they have a road map to those nodes does not mean they will be soon seen. The very simple reason is that those nodes cannot yet be produced by even Intel or IBM much less TSMC and the others.

Earlier this year both Intel and IBM announced making test circuits at the 10nm node and test structures at 7nm. If past history is any guide that puts test-production of functional main-stream use (eg CPU/GPU) 10nm chips at least 2 years out and 7nm as huge???
edit: looks like the past few months have been a surprise for 10nm node!

Not really.
You can either point several real miners at a pool using the same address or username and the pool reports it as 1 big miner or if using Avalons just point the one RasPi node at the pool and again attach as many 741's as needed to equal the supposed hash rate. Either way, the pool reports one user/address or supposed single rig as doing the work with no hint of the actual hardware pointed at that address.

The only reasonable non-hands-on proof is to see the CGminer display showing REAL SUSTAINED WORK. Not a 2-sec blip showing 70THs at zero diff. That is the systems sniffing each other to see how to work together. Display the CGminer header after several hours/days of running so we can see all the sustained stats. Sustained 70THs will be having a local work diff of around 50k or more forced on it by the pools. Sustain 70THs at that diff and then you may have something.

That would at least be a start. Can still be faked but judging from the computer/web skills so far displayed by scammers, well....

If a company wants to prove itself:
Display at some of the many Bitcoin tradeshows. Let it be seen, even in prototype format like Hotmine.io did. They demo'd running and near-finalized packaged miners. You know, real functioning hardware?  Hotmine built those 1st miners on their dime and publicly proved they are real.

They did not go straight to taking pre-orders for unseen/unproven tech much less use those same several years old recycled & stolen pictures of supposed Miracle Miners that are seen in every miner hardware scam over and over again.

After that - put demo units into the hands of known trusted reviewers for a month. Let them run/review it and then either return it or buy it from you.

That is how the likes of PandaMiner, Hotmine.io and Ebang went from highly suspected of scam to being at least reasonably trusted sources of hardware. They still need to build up a longer track record but all 3 are off to a very good start to being Reputable with a capital 'R'.

Well I do have that Intersil ISL8225MEVAL3Z 30A evaluation board I got a while back. Accepts 0V to 20V - yes it says zero - and delivers up to 30A with 0.6v to 6v output.

It is actually a dual 15A in one package - nice touch for multiple strings maybe? However, the eval board is setup as single paralleled output for the 30A.

Want it donated for ya to play with?