Topic: BTCFPGA bASIC updated from 54GH/s to 72HG/s (Read 10727 times)

Here for example: https://bitcointalksearch.org/topic/basic-btc-refunds-list-v20-152980

Those ordering from Tom are now screwed. Tom will not deliver and a lot of customers did not get a refund of the pre-order money. Tom is a crook and a liar. Forget doing business with Thomas Van Riper!

<3 ^^

The website is down where I can find info about this?

Do you have any source for this? I don't see why they couldn't be using UMC or ST, both offer 65nm.

lulz

Sounds like BFL when they started making their first FPGA ... we're experts at this for 10 years ...

They stopped saying that ... for a reason ...

Damn...... I thought I had managed to slip a private joke in without anyone noticing.
The plan was a post with the words "hand optimization"  & "In house custom ASIC design"

Hehe hand optimization, sounds dirty.

Is,n it a coincedence NON of the so called ASIC developers has delivered yet after screaming out loud for over half year yet delivering air over and over again gathering premoney ?? Even more if they are all using there own fabric methodes?  So or they are all fraud or they all suck really hard in promissing delivery date to naief well willing people...

TSMC is also being used by the Chinese company.
As for BFL "in-house design" I find that to be unlikely, they are not ASIC experts....Shit they are not even FPGA experts.
On a side note....
Just looking at some of the 'public domain' FPGA code floating about... Iv'e already taken sections of it from 160Mhz up to a tad under 300Mhz WITHOUT hand optimization......

So when I look around and see all the old FPGA's with 2 or 3 cores PER FPGA AND multiple FPGA's producing so few hashes
It makes me wonder.

You need to be more succinct  in your communication, If it is not going to impinge into your masturbation routine.

1. ASIC is a TECHNOLOGY that has been around for Decades, so it is highly unlikely it is a scam.

If you are referring to  bitcoin ASIC mining technology... then please state so.

Yeah, I believe it's(and someone correct me if I gaff) BFL = 65nm, bASIC = 90nm, Avalon = 110nm, ASICMiner = 130nm.

I hate the delay(duh), but I can't bail even though my orders are late in batch 1. I still have confidence that Tom can deliver the goods. I also believe he struck the most ideal balance between efficiency, maturity of process, and economics by going with 90nm. While BFL looks great on paper, going with 65nm for a first product was a tremendous risk and if their reports are to believed it's biting them right in the ass.  

Release a first product based on a proven design using a process that's relatively efficient and mature. THEN you can work on that fancy full custom 65nm hot rod in the back room and release that on the competition when sales wane. Intel type tick tock approach.

Everything I've read indicates that BFL used a small fab with their own inhouse design, while bASIC is using an off the shelf IP core with Bitcoin tweaks, manufactured by TSMC at 90nm.

But one is 65nm, and the other is 110nm? Seems unlikely.

I don´t think that is too far from reality. At least i wouldn´t be surprised. On the other hand i wouldn´t care either. So what ?

Due to a conflict of interest?....
There is a tasty tidbit...
I don't know what that means but very curious!
What if BFL and bASIC had the same companies producing thier chips!!!
Crazy? Yeah probably.

I love it when people argue points that were never made. I didn't say it was impossible. I don't see any of the three companies making a decent product anymore. They all seem to be in over their head. They could all go bankrupt trying to get something that works, and can be mass produced easily.

9/11 was done by BoA

_{(Bunch of Assholes)}

Very well put Meatball 

I feel for ALL who are delayed,not just BFL 

Yeah,all the ASIC companies are trying to delay the release of said products,WTF 

Talk about conspiracy theories,I bet they would believe 9/11 was done by BoA,because the mortgage was due on the World Trade Center 

ASIC done before. SHA done before. Sure, it's difficult and time consuming to produce the product in question, but anyone thinking it's impossible has no idea what the fuck they're talking about. If you think the idea of a bitcoin ASIC is a scam, you're goddamn delusional.

"ASICs are a scam" would be a stupid statement given that the technology itself is not new and is already in use.

The bigger question is whether Bitcoin mining ASICs are simply a more expensive shovel which can only be used for one thing - will anyone except the shovel-makers and the shovel-sellers ever make money from them?  And that's really something which buyers should have considered before placing orders.

........ you my friend need to read up on European law.
ATLEAST 3 YEARS, and bollox to them "not having spare parts".
 Apple recently got kicked in the nuts for their "extended" warranties
because under European law you have at least 3 years, so apple cannot legally charge for "extended warranties" during the warranty period.
And it has been this way for YEARS, just it is a big secret that these companies like to keep from their customers(for obvious reasons).

just because the different asic companies don't like each other doesn't mean they aren't scammers, nor does it mean they are competent enough to design something that never has been done before.

Ah, shit happens.  You can't blame Dave or Tom.  I can't imagine either of them isn't working their arse off to make things work and they probably haven't gotten much sleep lately.  Regardless of who your hoping to get your ASIC through, we're looking at brand new tech and there's bound to be bumps.

And for all the dinks ready to jump in with the "I told you ASIC was a Scam!"  This is for you...


Seriously, there's no way all these different vendors, some of which dislike each other a great deal, are in on some massive "Let's screw with people's with some vaporware!" plot.  We'll get there, it's just gonna take some time.

Ok, its official:
https://www.btcfpga.com/forum/index.php?topic=125.0

Maybe this:

https://www.btcfpga.com/forum/index.php?topic=115.0

Joke?

Link?

Posters on BFL forums are reporting having received email from bASIC giving a new shipping date of January 2013. 

well, problems with a webserver are not ultimately uncommon....

The btcfpga.com website is up and down today.  I wonder what's wrong.

It would have been a truly awesome move business wise if Tom was planning this all along. Now, people will have an easy choice for best bang for your buck ASIC  if you have decent KWh rates.

It seems like BFL is too late in the design/production phase to change their design like this. The only thing they will be able to do are firmware upgrades. Firmware upgrades for BFL to match bASIC speeds kind of sucks, because the bASIC will certainly have firmware upgrades of its own. I'd rather have the faster ASIC right out of the box.

A re-spin is a process whereby light sensitive material is reapplied to a previously processed wafer so a new masking can be applied and the wafer re-etched to correct bugs detected in a previous run.

......
Unless they foresaw a potential issue related to throughput and designed the PCB for 8 in the first place, and then once the Chinese started to bullshit, they thought  "fuck it" let's  fix  both competitors in one go......... and fully populate.....

What is "re-spin" and what are the implications?

That was my point, a lifetime warranty is effectively a one-year warranty, most likely. Still, it has to last that one year. If you start seeing mass RMA's in 3 months, you better roll your v.2 real fast after that and take trade-ins

I'd agree with that, though it would depend what you consider reasonable. I'd say something along the lines of twice the performance per dollar compared to whatever Gen1 is selling for at the time would be pretty stellar.

... actually I'd expect Gen2 to be a reasonable leap from Gen1
Gen1 is each companies first attempt at ASIC BTC mining ...
Though some of that leap may happen with the Gen1 hardware after release.

Did you read the quote I posted, from here? It was added by Josh a month after the initial announcement. This is the important part.
The first FPGA Singles hit the wild 9 months ago, and they now haven't been produced in almost a month. The SC Singles could be in production 50% longer and still the lifetime warranty would effectively be 1 year.

https://forums.butterflylabs.com/content/121-asic-products-will-carry-lifetime-warranty.html

You must have seen this...
and yes I agree that gen 2 won't be earth shattering either.

Do you have a link for that? I asked for clarification in the BFL FAQ about the warranty, but given the current wording it sounds like if the Gen1 Single SC isn't produced anymore and it's past one year the warranty won't be honored.

ASIC Gen 2 isn't going to be like the transition from FPGA to ASIC; it's likely not going to be a mandatory upgrade.

I'd rather have 54Gh/s three weeks before BFL ships than 72Gh/s three weeks after 

Still, any potential hashrate bump is a good thing, and these would be profitable at 1 billion difficulty so 

At which point you'll get your warranty filled and you can trade it in for gen 2.

From the BFL FAQ...

I would say that you cannot count on a warranty of any longer than 1 year. There's a pretty decent chance we might be in ASIC gen 2 a year from when BFL ships, and there's no real point in continuing the existing line once the next gen is ready.

Giving a lifetime warranty on a highly overclocked chip would be quite a gamble IMO. And since BFL does give a lifetime warranty for the boards, they must be pretty sure it can run non-stop for at least a year.

I wouldn't mind if it takes 80 or 100 Watt instead of 60, if performance rises accordingly. It would still be on top.

Pulling from page 1, because this is exactly what I was told (I added bold emphasis)

Man this is good news for me, I really don't mind if it delays slightly again but I hope before 2013 arrives.
For me getting the ASICs before the 25 BTC drop was never really a concern, I simply love the idea of
helping to produce an economy that's much different than what is currently available in this world.  I
hope the enthusiasm with the currency can stay positive until the big mainstream adoption. 

I agree, they should just do a batch revision in the interim after shipping what they now have and then offer an upgrade for people who bought the 60. Maybe upgrade them to 80 or 90 and then resell the 60's to other customers as a separate SKU.

It would suck to get a rig and one month later end up with a dead chip(s). That would be pretty hard to fix and there would be almost no workaround for that. (RMA's and the confidence hit wouldn't be worth it)

Overclocking a video card that's $200 and you have a huge amount of previous overclocking information from other people at your disposal is one thing.  Overclocking a few thousand dollars ASIC rig that noone has any idea how it'll handle that short or long term is a whole different ballgame.

Increasing the clock speed is the surest way to lose their power efficency advantage.

I know.. weeks ago I got deleted on butterfly forums when I asked if they re-spin.. I'm an Engineer specialized in this field, I can understand, mostly, the big picture. 65nm and full custom asic is difficult for a small company.. difficulty can be described mathematically.. Ofc BFL can be plaint lucky, but from customer POV it is gambling despite what they say.. I need to cover one - two months of gap, I need to buy something, I will use my skill to discriminate..

Give it another week or 2, won't matter much for the delivery date of those I think. Only be quick once one of them starts delivering, hell will break loose.

The question is, is the company over confident in a good way or a bad way?

They might choose a route where they take a hit in profit to provide extra hardware.
They might choose a route where they over clock their chips to their maximum rating and thereby increase RMA returns in March. ($$$ lost)
They might choose a route where they give away relatively cheap nick knacks...but this is unlikely to make most customers happy.

They might choose to upgrade their second batch and have an upgrade program (very likely and is the cheaper option). (Aka the 30, 60, 90 program?)

They might choose to offer a second device to cover the difference. (Possible, but cost ineffective)

etc.

Talking big in an overconfident tone might keep their customers around. (That's what real Vendors do anyway!) It may also just as well signal that they have no painless plan in place and need to talk big to cover for that inability.

One guy on their forums is about to switch his investments from BFL to bASIC at a cost to BFL of 8,000 dollars.

https://forums.butterflylabs.com/jalapeno-single-sc-support/427-basic-33%25-hashrate-increase-3.html#post5999

BFL is also now deleting threads from customers with content which is not unexpected nor even offensive considering they invested in the company as a paying customer. I just left a thread open in my browser this morning  and refreshed it....poof, it is gone.

The censoring has apparently begun on their forums (IMO).

Man, I totally agree with you, except one point, I need to buy more ASIC, atleast for next month until I shift, its gamble right now, God knows I don't gamble..

We will find out, but after delivery probably, by firmware update or whatever is used for it. They just don't want anything to stand in the way of getting the asic out the door asap. What a merry-go-round we're in. The "in theory capable of 16Gh/s" comes from the article in Bitcoin Magazine nov. 4th edition.

I am a bit lazy to search the post where BFL state that their asic chip are already overclock and that there is still a bit room left for further overclocking.
I might search it tomorrow when I am not lazy anymore.

How much room? I thought they can go to 200% but people say they are already overclocked.

Actually, bfl under-clocks the chips, in theory they can run at 16Gh/s, but stability and life expectancy would suffer. Just as btcfpga said for their chips btw. There is room for overclocking for sure.

Somehow, based on my experience, I think the same, but Inaba seems overconfident in last posts, I guess they might already got their "last version" of their chip, test it already and proved it qualifies.

more for the customers! weeeeee i think regardless people who pre-order will get their money worth. perhaps throw in a little single SC making a total of 90 gh/s

BFL already overclock the asics.
If they overclock to the maximum, the singles will do 8*10 GH/s=80 GH/s.
Not all asic chips will achieve that speed and I don't think they will last long.
Despite the smaller manufacturing proces they use, BFL asics seems to run slower per chip.
BFL chips are full custom made, I think they have plenty of room to optimize teir design at the next batch.
Cablepair is using already proven and optimize design technique. I don't think Cablepair has much room left to clock their chips higher if they want their chips to be reliable.
I think he will switch to a smaller proces at the next batch to stay competive.
Just my humble opinion.

If BFL will do that , that would be great. From the other hand, that could take more time. Wait and see.

Ditto... Paper hash rate wars in the environment where time is so much more important are just silly.

http://store.avalon-asic.com/

Avalon project has 9 left over from the first batch.

The only announcement from any company that will get me excited is someone announcing an actual shipment date. I could care less about a hashing power increase, especially when the rest of the announcement simply reads "we don't have any work done at all."

I'd have gladly taken a 20GH device 3 months ago over a 70GH device 2 months from now.

I hope each companie will deliver quality product, competition is good, but often compromise quality.  I prefer to have a solid, durable and reliable thing than a cheap one, whatever performance it accomplish !

Quote BFL_Josh: "We haven't decided specifically which way we want to make things even up with competitors products yet, but we have a lot of options and we will match competitors products that actually ship. We will not leave BFL customers behind the curve when it comes to hardware".

I hope BFL does NOT try to match this new speed. I'd rather get the new device in my hands first and let them worry about trying to increase the speed later when things have settled down. But competition is a good thing. Congratulations.

And so the saga goes on..

So now we just wait until BFL comes with a (40) (60) 80gh device..

Maybe they will get some more pre-orders now..and gives m more time(to)frame..

How much of the board design work and time was wasted by changing to 8 chips per cluster? Haven't they been working on a design with 6 chips per cluster for many weeks? How much has to be scrapped?

As far as I understand it, all 3 major vendors first batch pre-order lists are full. Always room on the 2nd wave lists though (think they said that's scheduled for Jan '13)

Wow this is great news and a smart move. BFL has better hash/watt ratio but bASIC now has a much better hash/$ ratio. If BFL delays again I think Tom will pick up some diversifying miners.

Is there still room on the preorder list?

The hardware wars begin!

I mean I'm wondering what he's done?

No offense Easy2Mine but as much as i'd love 2 Singles for the price of one this isn't going to happen.  You should read up a bit more on BFL's actual process and I think you'd realize that BFL shouldn't have an issue bumping up the speeds of their chips to match this...  Currently at 7.5GH per chip at 500Mhz and able to generally reach 1Ghz (if there was proper cooling).  *From the information that's been given I think it'll just take a firmware tweak to bump it up no problem.

I'm certainly no fanboy but I'm quite sure it's not an issue.  That said if I can find a way to come up with some more $$ I'd like to pick up a few of Toms bASICs asap.  Tom is using the 90nm manufacturing process for his ASICs and will generally use more electricity than BFLs 65nm process thus giving off more heat (in general) but, I think he's doing pretty damn well with what he has, and will be keeping ever other ASIC Company on their respective Toes.  I hope we all end up getting more GH/$ because of this as well as keeping our Electric bills a little more reasonable in the near future.

Now we just have to keep waiting as there're more delays from both companies because of changes made.  I hope they are minor and finish up sooner than later...  

BFL single has 8 asic chips per cluster.
Their boards and enclosure are already at final stage.
I don't think BFL has enough room to overclock the chips within a safe working enviroment.
The easiest way is give every customers 2 BFL singles for the price of 1.
2 BFL singles*8 asic chips per board=16 asic chips
Same amount chips as the bigger bASIC unit and we are all happy.

The smaller unit is upgraded from 6 asic chips in a cluster to 8 asic chips for a total of 8*4.5 GH/s=36 GH/s
The bigger unit has 2 clusters of 8 asic chips. 2*8*4.5 GH/s=72 GH/s

Each BTCFPGA asic chip can do a maximum of 14 GH/s, but it won't last long at that speed.
6 chips per cluster are more difficult to implement than 8 chips per cluster.
I personnally think that they took the advice of the board designing company to change it to a 8 cluster or 2*8 cluster board.
Cablepair has already say that he is hiring one of the best board designing company to do the job.
In my humble opinion, I think they will have a greater chance to deliver on time if they go for the 8 chips per cluster board instead the 6 chips per cluster board.
But who am I?
Only time will tell.

Wow, a 33 percent increase over the previously stated speed on this yet-to-be-seen product?   That's fast!

what?

luke-jr?

aside from shitting on other people that disagrees with him.

Eh what?

8 chips x 9gh/s = 72.

Wait, so their chips were running ~14GH/s, with the cheaper unit running 2 chips (2 x 13.5 = 27), and the more expensive unit running 4 of these (4 x 13.5 = 54). They found out these were clocked too high, so they had to decrease performance. To compensate, they added an extra chip to each module. How many chips is it now? 3/6? 3 chips would sound about right (3 x 12 = 26) for the smaller one, and 6 chips for the larger one (6 x 12 = 72).

A) So they added 50% more chips, but only decreased the clock rate by ~10$, and somehow this is expected to use about the same amount of power? That doesn't make sense.

B) In order to do this, someone else mentioned that they have had to redesign their board, and redevelop their firmware (which may or may not be even completed yet), which just adds time. They're not even close to assembling hardware, are they?

C) A 50% increase in the chips used per unit means that whatever initial batch of chips they ordered, they can only ship 2/3 the same number of products. If they ordered 3,000 chips, they just went from being able to ship 750 units to 500 units.

D) I am loving the increase in hash speed at the same price point. Competition at it's best!

it'll most likely delay the production though.

well, there goes beginning of december..here's hoping close to the end of december.

Not sure how many in this thread read the post referenced by OP, as most people are seeing "ASIC war!" but it looks like this is simply a result of ASIC stability.


Might be BS, or might be true, that it's simply a matter of the fact that 8 is more stable than 6, and that's why they chose to go this route, which will end up with some happy customers. Rather than being anything to do with outdoing BFL (though that doesn't hurt).

EDIT: Reminder though that chips were apparently literally dying at overclock speeds, so before you have wet dreams of overclocking, keep in mind the horrors of murdering your ASIC chip.

and that's still a conservative estimate.

so it might be even faster than 72 gh/s

in normal circumstances, i strongly dislike luke-jr, but at times, he can work some magic.

BFL's ASIC can match that if they want. At 40Ghash their ASIC was running at 500Mhz, and they said it can do 1Ghz but not all chips will reach it. So 72GHash is possible.

2 words:
FUCK YES.

3 words.
SUCK IT, BFL

Competition is good. Everybody wins.

However, this announcement essentially includes the statement "we actually have nothing done whatsoever." So until both companies actually deliver, there will be no benefits.

Heh, how great is it that there's a 'performance war' between companies before anything has even been released?  I imagine each ASIC vendor sees a rush of pre-orders every time one of these announcements come out.

HOLY SHIT. I support you BFL, but this is bad news for us BFL supporters.

Yes, but with a delay.

Hmm, so they are selling 72GH/sec for $1069.99 now?

BFL also wait for news Tuesday, their last chip version coming from fab.

I'm a customer as well, I'm not sure what to think on the timeline,


Hopefully there will be good news Tuesday.


Has BFL mentioned anything about it's shipping schedule?

But most likely this will add 3 extra weeks to deliver. One week redesigning board layout to accommodate more chips (this includes power circuit redesign), one week (they said 8 days) to produce pcb, one week to assemble everything.

PS: I'm a customer.

From https://www.btcfpga.com/forum/

Those ordering from Tom are happy!-)