No announcement for AB was due to NDA's with BDPL stuff. Overt and/or covert AB requires it to be implemented in hardware. Putting circuitry that you may or may not use in an ASIC kinda defeats the purpose of an asic (ie stripped down processor to do one explicit task). Anything else is wasted space. You cant use AB efficiency claims numbers and just add it to what you think an asic should hash at. Especially if you are talking about a new processor node size. I have no more information than the rest of the internet. I just know in R&D, pushing the technological envelope will always require some fine tuning to eek out the most performance gains. In just a few weeks, ck has been working on firmware and has improved efficiency by a pretty good margin. It takes miners in hands and a good sample size to get any product vetted. This isnt an iphone or other mature industry product and thus I wouldn't expect the hardware to be fully flushed out before getting into hands. would never get anything out before major diff changes or node changes happen. Since announcement network has what, almost doubled?
So your option in this area is......A) Let mfg mine the shit out of the coin (6-9 months) to tweak and tweak before shipping, or B) start shipping miners when you are damn close and then let the community help stabilize the product.
We are talking about 10% deviation here and I'd bet much of that is just difference in power quality from either power supply or mains as can be seen by testing all the different supplies..... if your profit of make or break is reliant on that 10%, then I think your model needs adjustment. Better to take whats available and run today and profit rather than bitch and moan about a few points to the bottom line.
1. then they had 0 bizz sellin miners to customers. the limitations of these miners is a HUGE deal my dude. u may not think so but remember when folks thought how antminer s9's could only mine on antpool?? do u recall how freaked out everyone was by this?? it turned out to be false but that is what happened here with the t1 and because of it no one can mine ne thing but btc on 2 pools with these overpriced machines. u better hope nothin happens to asicboost otherwise these machines cant be used period. maybe ur fine with undisclosed limitations on hardware u paid a premium for but others r not.
2. we r talkin about a 100% deviation not 10%. they require btc and asicboost....nothin else will work. u cant even take a 20% loss of hashrate and go elsewhere with ur machine because like others have said....they will not work at all period non 0 zilch....nothingggggggggggg. that IS NOT what folks paid for. in the words of phil....prove otherwise and i will retract my statement.
3. the info given was based on their miner and their PSU. if u have to spend another $300+ and drop the hashrate down by 2th/s to hit the numbers that claimed is that not somethin u should care about?? u can claim a 10% difference but ur not factorin in the cost of the beast;y psu phil had to buy along with the decreased hashrate that comes from matchin the power specs they promised. or maybe ur alright with payin the premium and gettin less than what was promised. if so carry on my friend lol.
a user "bip37" claims the asic and hardware for the s9 in general supports overt and covert asicboost it just needs pool lvl support and possibly firmware support to configure it properly since the s9 by default (on our end at least) has asicboost turned off. he also goes into detail about how to sort of turn it on without doing any custom this and that if u want to test it out.
dude!!!!! so your're all basing your information on a post by a user on a forum? lmao. bro, btctalk is not due diligence, very few posts, including ones from myself represent nothing other than most people opinion, and unless info is coming from an inside engineer that is dick deep in one of the ASIC companies, I wouldn't hold a candle to anything anyone says. Holy shit, I had no idea I was trying to talk sense into some info gleamed from a posting lol.
4. he posted information and a link to bitmains source code for their miners. this is what they posted....r u sayin bip37 isnt trusted or worth trusting?? r u sayin the information he posted isnt and hasnt been looked into by other members?? this info hasnt been proven wrong yet u claim its wrong?? i guess my next question is who r u?? 50 posts and joined a few months ago...what makes ur assessment viable?? have u tried verifying the claims?? what were ur findings?? why was no information provided when u replied to attempt to insult me??
if ur basin ur comments off of engineers or spokes poeple for the company y r u not talkin about the fact that halong withheld the asicboost limitations from customers?? u can claim nda but a NDA doesnt take blame away from ur company for false advertising. do u understand that lies r bad?? do u get that most basic of basic ideas??
also i should add many members of this forum who r not related to asic companies at all have posted guides and info that allow u to maximize the efficiency of multiple model miners including but not limited to the s7 and s5....just to name a couple. now since they arent engineers with bitmain ur comment implies that they r sellin snake oil and r hurting customers in the process. im pretty sure they wouldnt take kindly to folks spwein these kinds of comments that end up groupin them into whatever shade ur tryin to throw here. the moral here is if udk ne thing about what ur sayin or dont bother doin it urself dont comment tryin to piggy back on what another member says. phil can claim whatever he wants...hes an adult....he doesnt need u to jump in flyin off half cocked lookin silly. info starts somewhere. rarely does it ever leak the way u want it to. i guess its good that ur not known or trusted here....we may never be able to pass information to one another if u had ne say about it. afterall isnt that what this site is supposed to be for?? silly kid.
)