Topic: Swedish ASIC miner company kncminer.com - page 1828. (Read 3049501 times)
Seems I have to stalk your profile for latest posts then
Mind making a thread comparing KnC, HF, BF, Avalon, Labcoin, BTCGarden, Cointerra, xCrowd etc?
To have your comments somewhere concentrated, you seem very knowledgable..
Thanks for the compliment.To have your comments somewhere concentrated, you seem very knowledgable..
Unfortunately, there isn't sufficient technical information available to do a meaningful engineering comparison.
The only meaningfull contribution I can make is to shot down impostors who use the correct English grammar and technical vocabulary, but have no actual understanding of the underlying technology, e.g. kingcoin.
KnC had published some general technical information couple of days ago: https://www.kncminer.com/news/news-25 . I can't access it now, but it clearly showed that the package has over 1000 pins and that the single chip is comprised of 4 disconnected sub-regions, with absolutely every feature symmetrically quadrupled. Each of the quad in turn contains 24 identical copies of the hashing engine. When you can access it take a look at the floorplan and make a guess: what would be the probability of randomly distributed defects to hit every of the 4 copies of "control logic". What would be a minimum required size of a "single defect" that would kill all 4 regions? Even if you aren't rocket surgeon or brain scientist your honest guess will be better than the kingcoin's assesment of "insanity" of the KnC designers.
Here is a fragment of the Wikipedia definition of a code monkey: a derogatory term for an unskilled programmer who is only able to perform trivial or repetitive computer programming tasks or a reference to a job that treats even experienced computer programmers in a way that trivializes their problem solving abilities.
CAD-monkey is the hardware equivalent of the above term, because most of the hardware engineer's work involves not writing/editing "code", but operating some "CAD" or "EDA" tool/program.
Edit: CAD-monkey is derogatory if the person claims to be a qualified engineer. It may be a very friendly, or self-depreciating, reference to a skilled/trained salesperson of the CAD tools. Working in sales could really require skill and patience at repeatedly explaining/presenting "how to import a netlist" or "how to export to an ATPG (Automated Test-Pattern Generation) tool".
A stuck at fault in the control logic or in the IO interface.
You are incompetent.
The above is the proof.
Hey hey, no reason to get rude here!
You should have a look at their R&D slides again. Their die contains of 4 fully self contained "quads" , 48 unrolled hash cores each. Each core must run at about 520 MHz to realize in sum 100 GH/s.
The complete die will have a size of at least 100 mm2. It's 28nm, not 130nm, there will be yield issue for sure for such a huge die. Most likely this will be stuck-at faults in some of the hash cores (which could be easily detected by a BIST or ATPG tests). If these permanent production failures cause permanent non-working hash cores, the remaining cores have to run at a higher frequency to still have 100 GH/s per chip. Because this is what they have to sell.
If 3 of the 4 quads have stuck-at faults in the interface logic, it would be impossible to realize 100 GH/s with the remaining one quad, because in this case it has to run at about 2 GHz.
Of course one could try to implement each quad to be able to run at 2 GHz speed too, but this kind of "margin" will probably double the area you need for one hash core. I doubt that they have added enough margin to compensate even one complete non-working quad to still have 100 GH/s.
The skipped traditional production tests will force them to realize all this compensation on-line in the final miner by firmware procedures. This it not impossible, but the development of that feature is not that straight forward, because they currently do not know how a higher number of chips will behave in terms of yield and how these procedures have to look in detail.
With regards to tracking an idea comes to mind. I was planning on putting a gprs modem in the router for the racks of machines to act as a backup internet connection. $40 USD for prepaid sim card that charges $2 USD a day when active and $60 for the modem. But the modem also has GPS. One could easily have it not only fail over to the GPRS but also beacon the GPS location.
Again, I'm not worried about theft from a data center.
That's a good idea. Are you suggesting something like this? http://www.ebay.com/itm/Sprint-Compass-High-Speed-597U-597-USB-3G-GPS-Mobile-Modem-AirCard-FAST-SHIPPING-/221217870969Again, I'm not worried about theft from a data center.
A stuck at fault in the control logic or in the IO interface.
You are incompetent.
The above is the proof.
To have your comments somewhere concentrated, you seem very knowledgable..
A stuck at fault in the control logic or in the IO interface.
You are incompetent.
The above is the proof.
There is probably some logic selecting which core(s) to use and some common IO interface. A single fault here could render the device useless depending upon the design. Give me the netlist and I'll show you where the possible faults are.
A stuck at fault in the control logic or in the IO interface.
You are incompetent.
The above is the proof.
That's insane. If something is wrong they will not know if it's a problem with the ASIC or not. They can't complain to the ASIC vendor either and they will not have a clue in terms of yield. Did they do this to save time so they would not have to learn and run ATPG tools?
What are you talking about? What would be an example defect that would completely disable a chip with 4 non-overlapping/non-intersecting clock trees?A stuck at fault in the control logic or in the IO interface.
Hehehehe, Hook, line, & sinker 
What are those mysterious dark shadows inside the case on the right?...
Is that possibly a pcb I see hiding in there?
Just to "Stirr the Pot" a bit...
lol, get your eyes checked old man! hahahaha
What are those mysterious dark shadows inside the case on the right?...
Is that possibly a pcb I see hiding in there?
Just to "Stirr the Pot" a bit...
No that's just scratched metal surface. The case on the right is assembled.
What are those mysterious dark shadows inside the case on the right?...
Is that possibly a pcb I see hiding in there?
Just to "Stirr the Pot" a bit...
Sadly the biggest worry on shipment would be customs if anyone has been following the Avalon chip hostage crisis...
I would be driving to the customs office pronto on first hint of any delays
Feels so good to be in the EU right now and so close to Sweden. I would expect to have a 1 day delivery straight to my door
Well that's exactly what I just had happen a few weeks ago. Unless the size makes a big difference. I lost a gold card and reader for Entropia and it took a day to replace by courier from Sweden.
I'm never going to believe that all rigs will be hashing with an hour of completion however..especially the first ones. How many a day are they producing? Got to be 100 or more which will need to be transported (and packed and unpacked for that), be wired up and have miner and pool details sorted (accurately). Granted they won't just have one guy doing it ...but I doubt they have Ooompahloompahs either.
"Here you go boys, the 1st batch of 100 is ready to set up, see you in an hour when the next batch is ready" lol
That's insane. If something is wrong they will not know if it's a problem with the ASIC or not. They can't complain to the ASIC vendor either and they will not have a clue in terms of yield. Did they do this to save time so they would not have to learn and run ATPG tools?
What are you talking about? What would be an example defect that would completely disable a chip with 4 non-overlapping/non-intersecting clock trees?Please go back to your ATPG tools and obtain the probabilities. Then report back here and re-state who is insane.
Also: use your best design-for-test methodology to create a test pattern. Then compare the coverage with just any 10 random 80-byte test patterns. Then come back here and report who is a CAD-monkey in this thread.
Edit: I'll add this explanation for the readers unfamiliar with logic design. SHA-256 is essentially self testing: there's no internal state that is either not controllable or not observable. Same with all the interconnects. It is possible to construct a very-low-probability faults that will go undetected in the unrolled design. But their probability is way lower than the acceptable hardware error nonce rate in a Bitcoin miner (single percentage points).
For the readers unfamiliar with logic design but capable of compiling a program: take an example working SHA-256 code and make a single character change somewhere in the code, e.g. '0' to '1'. Then recompile and compare the results of hashing a test file with 1000 randomly choosen 80-byte strings. You'll immediately understand what it means "high toggle rate" and "self-testing logic".
This self-testing property has been discussed on this forum multiple times, in English and in other languages.
That is true. The timing models and static timing analysis of the ASIC's are very accurate. But they are given at a certain power supply level and temperature. If you mess up your mechanical, thermal and PSU design so these conditions are not satisfied then you're pretty much screwed.
However, if you then start to test your system and push the margins you will be screwed again since you don't know where in the process window the particular chip you have in the lab is. If it's a best case device then your systems might fail when you get a new batch of ASIC's which are not best case. Your production line will halt and you have to start debugging again. Or even worse, if your production test is not optimal you will not notice until customers starts to complain and return your product.
However, if you then start to test your system and push the margins you will be screwed again since you don't know where in the process window the particular chip you have in the lab is. If it's a best case device then your systems might fail when you get a new batch of ASIC's which are not best case. Your production line will halt and you have to start debugging again. Or even worse, if your production test is not optimal you will not notice until customers starts to complain and return your product.
There will be for sure a lot of fun in the KnC lab!
They stated at the open day, that they will not do any classical ASIC production tests (neither wafer test nor final test). If this is still true, it means they will package the dies and put the resulting chips on the PCBs blind.
If the chips work at all and at which speed they probably can determine by running a complete miner only. In worst case this can result in tweaking each box individually.
That's insane. If something is wrong they will not know if it's a problem with the ASIC or not. They can't complain to the ASIC vendor either and they will not have a clue in terms of yield. Did they do this to save time so they would not have to learn and run ATPG tools?
You've been one of the few that followed through and made some calls prior to make sure you're behaving responsibly before purchasing anything here, so good luck bud
When I spoke with him today he made a remark: "It's good you did a research", which most people don't do
Still I don't see a reason why you souldn't register as a reseller
It's free
I may do once chips are in their hands and proven to exist. Until which point I'm still the scum of the earth according to this forum. I'd be damned if I did, and damned if I didn't. I'd personally prefer the lesser of two evils and wait before adding any kind of reseller link for myself. The abuse is just not worth it. I'm sure if everything work out as planned there will still be reasons to demonise me. I actually kind of regret going if I'm honest.
Dude, If they ship on time I'll buy 10 jupiters through you in november. Lets just hope they get everything moving for an early september release
O'rama... Yeah, I know, but with the 4000+ people going to look for PSU's at once... One could end up in "A pickle" if the ATX's end up "out of stock" for a bit.
Thousands of people world-wide, placed over a period of a weeks - that's not going to be a problem for even one store to handle and restock. Unless there's an earthquake that takes out manufacturers in China/Taiwan/Korea then it won't be a problem.
Yes, disturbingly it may involve standing, walking and driving to a store over a mouse click.
I'm actually stripping old systems as well, for the psu's.
O'rama... Yeah, I know, but with the 4000+ people going to look for PSU's at once... One could end up in "A pickle" if the ATX's end up "out of stock" for a bit.
Thousands of people world-wide, placed over a period of a weeks - that's not going to be a problem for even one store to handle and restock. Unless there's an earthquake that takes out manufacturers in China/Taiwan/Korea then it won't be a problem.
Yes, disturbingly it may involve standing, walking and driving to a store over a mouse click.
Some of us ordered enough miners that it is unlikely a single store location would have enough quantity in stock... Further I would be disappointed if their DC to DC converters were not tolerant of ATX 2.3 spec supply ranges.
O'rama... Yeah, I know, but with the 4000+ people going to look for PSU's at once... One could end up in "A pickle" if the ATX's end up "out of stock" for a bit.
Thousands of people world-wide, placed over a period of a weeks - that's not going to be a problem for even one store to handle and restock. Unless there's an earthquake that takes out manufacturers in China/Taiwan/Korea then it won't be a problem.
Yes, disturbingly it may involve standing, walking and driving to a store over a mouse click.
O'rama... Yeah, I know, but with the 4000+ people going to look for PSU's at once... One could end up in "A pickle" if the ATX's end up "out of stock" for a bit.
Thousands of people world-wide, placed over a period of a weeks - that's not going to be a problem for even one store to handle and restock. Unless there's an earthquake that takes out manufacturers in China/Taiwan/Korea then it won't be a problem.
Jump to: