Topic: HashFast announces specs for new ASIC: 400GH/s - page 535. (Read 880461 times)

Datasheet says 38.35mm and the fets are 0.7mm nominal

Main aim was to get the down stream one a good distance above the water pump.

Contact [email protected], they should be able to help you.

We are preparing a special promotion just for batch 1 baby jet purchasers. Coming soon.

Yes - chips will be available for developers. Please contact [email protected] for more info until we get them on the website.

To John S.,

In my PM to you yesterday morning on behalf of my co-operative, I'm sorry if I pissed you off John (something I seem to do if your name is Yifu as well) but at least the Flash Sale worked for me and thomas_s. to reserve 100 HashFast Sierra shares worth over $11K on the GB forum in less than 12 hours. That may be a GB forum record if any others would like to dispute that.  

So, we're seeing if we can sell a 2nd Sierra, this one cheaper, without MPP, to try to beat the Monday night deadline: https://bitcointalk.org/index.php?topic=299171.0;all

Please PM me if you have questions or would like to take a fractional risk on what are looking like the best available ASIC pre-order deals at the moment.

So with this miner protection program? the motherboard supports 3 HashFast Golden Nonce (GN) 28nm ASIC chip? and what is gonna cost if we have to add two more HashFast Golden Nonce (GN) 28nm ASIC chip?
if we dont get RIO, because today I looked on http://mining.thegenesisblock.com/ and the break even is 96 days and most likely gonna increase? So does that mean on day 90 I get 2 more HashFast Golden Nonce (GN) 28nm ASIC chip per Babyjet?

For those who already ordered a first batch Baby Jet, would it be possible to trade it for a Sierra on sale? How much extra would that cost for the device (beyond the $400 difference) and shipping?

You may be surprised how unimportant hashrate may be a month later. I'm in group buy for batch 1# BabyJet and I would probably refuse to exchange for 3X hashrate a month later. Given the circumstances of KNC and Bitfury unloading miners in that period it may actually return less BTC. Current monthly difficulty increase is estimated at 115%.

Well a month later is a big deal.  Expect prices from all vendors to keep falling though. 

As someone that's ordered a batch 1 babyjet, this price drop seems like a huge blow to initial customers that believed in Hashfast.  The Sierra is offering 3x the hashrate for almost the same price.  I know there is going to be the MPP, but at this point, one can just get 3x the hashrate only a month later than the first batch.

For the next 3 days, prices have been slashed on Baby Jet’s and Sierra’s. Now is the time to invest in additional rigs. Sale ends Monday at 11:59PM PST!

http://decentralizedhashing.com/2013/09/hashfast-tapeout-sierra-mpp-new-next-gen-chip/

Will chip be available for developers?

Thanks for the update and I am glad you confirmed this detail.  The entire package is 45mm by 45mm, is the lid smaller?   If so can you provide the lid dimensions?  
If lid occupies entire chip size that would be a surface area of 20.25 cm and a heat flux of 12.3 W/cm^2.

For those wondering what I mean.  Two example chip packages, pretend each is 45mm x 45mm.  The surface area of the lid for the  chip on the right is significantly larger.



Good to hear.  What is the operating temp the software targets? 60C?



Glad to hear that.  Care to share any details on radiator size?  Will all fittings be PC water cooling standard (G/14 threads, 3/8 or 1/2 ID barbs/compression fittings) to make upgrading radiator and other components possible with off the shelf components.


Thank you.  So I see it something like this.

Chip power: 250W  (~333A @ 0.75VDC)
Board power:  284W (~24A @ 12VDC)
System DC power: 315W (~@27@ 12VDC 284W board + 31W for controller, fans, pump, etc)
System AC power:  350W  (~3A @ 120VAC assumes 90% ATX PSU efficiency)



Thanks.  Eyeballing the diagram I assume the case is 2U high.  That would make the fans 80mm.  If diagram is to scale that puts the FET heat sinks @ ~60mm (+/-10%).  Are FET heatsinks removable (out of warranty of course).  The heat flux should be < 1 W/cm^2 which is trivial in immersion cooling.  The "shorter" the board height the closer they can be spaced and the less very expensive Fluorinert I need to buy.  Also some users may wish to replace them with water blocks.

 Of course and understood.


Excellent.  Thank you for all the info/details.  Keep it coming.  

Hashfast an idea, how to cool the radiators on your watercooling: http://youtu.be/VUbpb23yTK8?t=11m6s

AHa. That explains a lot. I already wondered why anyone would develop such a large die for bitcoin mining.
It really is only a ~100GH/s 81mm² die, with 4 of them in an MCM. Interesting, and probably a good idea.

thanks for the indepth info, Simon.
beautiful design so far, cant wait for the final design.

Anyway, on the design im working on, i need the height (approx is fine) of the two conventional heatsinks you mention, let me know when you have them?

ttt

Overall, our chip most closely resembles a modern "hot CPU" design (Intel Sandy Bridge E, AMD FX-9, SPARC T4, etc.). The package is a BGA, 45mm x 45mm external dimensions. The total die area is approximately 324mm^2. The die is split in 4 - i.e. under the metal lid there are 4 dies in a square arrangement, each approx 9mm x 9mm with 5mm gap in between each one (for better heat dissipation and spreading).

One highly relevant feature regarding power use is that the GN chip incorporates on-die temperature sensors and a control system designed to adjust voltage and clock speed to the capacity of the cooling system.  Thus if the cooling system can dissipate a greater amount of heat, the software can "overclock" the chip to fit it's power usage to the heat dissipation capacity, and produce greater hashing capacity.

Similarly, the chip will "underclock" itself in response to external circumstances that reduce the available heat dissipation capacity (say, a very hot day, a failure in the cooling system, a blocked air vent...).  The overall design intent is for the chip to always operate at the maximum possible hashrate dictated by the circumstances.

Finally, one recent piece of news is that we have received results for the stage-III thermal test (full physical prototype) from our cooling system partner.  Stage III tests involve the creation of a full and complete prototype of the system, including the same case, fans and cooling system that will be used in the Baby Jet as shipped.   The only difference is that the chip is substituted by a variable-output heating element with the same form factor as chip. The test consists of running the Baby Jet as if in production, and increasing the wattage produced by the heating element while monitoring temperatures.

At the nominal operating point (400Gh/s) the best silicon will consume ~250W according to our simulations. There is some variation in silicon however, so some silicon will consume a few % more. This power level is at the chip only. The system has 2 power conversion stages between the wall plug and the chip - first an ATX power supply that outputs 12v. This supply is about 88% efficient. Then there is a second supply stage on the module board the chip is mounted on. This second supply stage converts the 12v down to approximately 0.7-0.8v that the chip runs at. The combination of the losses in both PSU stages and the additional consumption from the pump, fans, controller etc account for the difference between the 250W at the chip and about 350W at the wall.

The module board is ~4" wide and approx 10" long (may end up a little longer, up to 12" - we are still configuring the power connectors). Here is a draft layout - note this is not final and is subject to change. Dimensions in mm.



The tallest board components are two conventional air cooled heatsinks on the FETs that form part of the power supply. I don't have the height to hand - I will get back to you. Here is a rendering of a draft of the module boards, installed in a Sierra. Note - this design is not final, and is subject to change.



Yes - the controller (which is a raspberry pi) should be able to control many boards. We have not analysed how many. Our chip has been designed with a particularly advanced internal interface, and supports variable difficulty levels and n-time rolling on chip. This massively reduces the traffic from the controller to the chip, keeping the CPU requirements on the controller very low. The interface supports daisy chaining up to 63 chips to a single serial port on the controller. The traffic levels and limits will depend on the difficulty levels the chip is configured to run at, and the use of n-time rolling.

Yes - we plan to add complete module boards to our website shortly. These can be directly controlled via either a serial port, or USB. We will be open sourcing the drivers in CGMiner.

Not sure who else saw this or not but on the baby jet batch one it now says

"The Baby Jet motherboard has room for an additional GN chip module.**"

the 2 stars refer to simply that you need an upgraded water cooling system and upgraded power.

At this point tho from calculations i have run it doesnt look like anyone can really breakeven tho

Didn't miss it... Sent this earlier to my Engineering Team for review.