[Group Buy] Avalon ASIC Chips (SebastianJu) Batch 6 ordered (Closed) - page 37.

ajas

member

Activity: 130

Merit: 58

Quote from: SebastianJu on June 25, 2013, 01:28:56 PM

Look at this info from Bitsyncom/Yifu:

......

But i find it funny that the old technique, that is used by Asicminer or Avalon is nearly as good pricewise like the newest asic-tech that comes on the market.

Only some thoughts. Correct me if im wrong.

KNCMiner says: 500 W 175 GH/s that means: 2.86 J/GH
Burnin/Avalon says: 40 W 5.56 GH/s that means: 7.19 J/GH

So on the long term, when difficulty increases and power requirements count, the new techniques should be better.

But nobody knows if KNCMiner meets its power requirements or will ever deliver.
However we are pretty sure that we will get something from Burnin/Avalon in time.

All ASIC developers (AVALON, BFL, KNCMiner) try to sell their products at prices as high as they can.
Much more than the actual costs.

SebastianJu

legendary

Activity: 2674

Merit: 1083

Legendary Escrow Service - Tip Jar in Profile

Quote from: ujka on June 25, 2013, 02:07:43 PM

450 Mh will need some elaborate cooling, but yes, I agree that those chips have great potential, even compared to some newer chips (promised chips Wink

Especially for DIY people that want (need) to fiddle with hadrware. I don't see where is the fun in buying one of those 'black box' miners? For quick protif only?

I wonder what cooling it needs. Its not a gpu that is running at high heat anyway, its a complete new type of chip whose overclocking capabilities isnt tested yet really. So i can imagine that it might be even possible with proper airflow and heatsink to cool it down good enough. I mean didnt someone post fotos in the forum that the heatsink from avalon miners doesnt even has contact with most of the asic-chip-vias? That means these chips arent even cooled by the heatsink. So i guess the new designs could hold some surprises...

ujka

hero member

Activity: 630

Merit: 500

450 Mh will need some elaborate cooling, but yes, I agree that those chips have great potential, even compared to some newer chips (promised chips Wink

Especially for DIY people that want (need) to fiddle with hadrware. I don't see where is the fun in buying one of those 'black box' miners? For quick protif only?

SebastianJu

legendary

Activity: 2674

Merit: 1083

Legendary Escrow Service - Tip Jar in Profile

Look at this info from Bitsyncom/Yifu:

Quote from: BitSyncom on June 24, 2013, 11:11:37 PM

was wondering how long it'd take people to notice ( and more importantly share the constant that we've released on github.)

the number you are all aiming for is 450 Tongue

of course, that's not really possible on just air cooling.

He is saying that avalon chips have a clockcap at 450MHz. They now run at 282MHz. That means overclocking potential of 62.6666%.

I calculated before, the Jupiter Miner of KNCMiners cost around the half of the price you would pay when buying Avalon Asic Chips and buy the assembly for them. At least the cost per GH.

Now burnin and BKKCoins are both trying to revise their design to use this overclocking capabilities. That means... even when KNCMiner, or Bitfury, that has similar prices per GH like kncminer, are cheaper now their advantage is gone completely. At least when taking into account that their miners are delivered months after the chip-avalons are mining already. BFL's credits arent worth anything anymore in my opionion too.

Even when i thought a moment about if it would be wise to invest in KNCMiner i dont think so now anymore. Of course the chips of kncminer, bitfury or bfl has to show their overclocking potential too.

But i find it funny that the old technique, that is used by Asicminer or Avalon is nearly as good pricewise like the newest asic-tech that comes on the market.

Only some thoughts. Correct me if im wrong.

SebastianJu

legendary

Activity: 2674

Merit: 1083

Legendary Escrow Service - Tip Jar in Profile

Quote from: daemondazz on June 24, 2013, 11:22:02 PM

First sample chip arrived today - well, at least I'm assuming it is, the wife sent a photo of a Deutsch Post parcel and I'm not expecting anything else from Germany!

I hope you can make good use of it... :9

daemondazz

sr. member

Activity: 448

Merit: 250

First sample chip arrived today - well, at least I'm assuming it is, the wife sent a photo of a Deutsch Post parcel and I'm not expecting anything else from Germany!

lawnmoverman

newbie

Activity: 5

Merit: 0

; <17>; <1.462>; <1NLHTKFVrLEursfqYxcxGxqEiRGTeCiG7E>

idee2013

sr. member

Activity: 407

Merit: 250

Have to sell my avalon chips from batch 1 and 2 https://bitcointalksearch.org/topic/auction-591-avalon-chips-from-group-buy-to-sell-ordered-in-may-with-escrow-242267

SebastianJu

legendary

Activity: 2674

Merit: 1083

Legendary Escrow Service - Tip Jar in Profile

Just to inform you: I sent 2 samples to strombom and his project since he didnt get samples from zefir yet.

Schrankwand

full member

Activity: 224

Merit: 100

Quote from: svetliopi44 on June 23, 2013, 02:02:55 AM

I want to thank my colleague ujka . But are these 9% accurate? In my experience the difficulty increases by more than 20% of any further increase.
Please tell me where miscount, but my calculations that comes after five months would be pointless to have such a machine of 320 chips.
Please discuss, I'm new in this area and obviously wrong somewhere.
Thank you.
Sorry for bad english.

It is completely useless, from my personal understanding, to use any constant percentage increase. It is more interesting to calculate a certain total hashrate that is probable from all existing source of Bitcoin mining power and from there create a certain target difficulty.

When the avalon Chips hit. You will see a few more 25% increases, but from then, even BFL shipping hardcore will not make many of these increases, since we are talking a compounding interest system.

20% for 12 months would mean 1,2^12, coming up to an increase of 8 times current network power. Yes, that is possible, but at first unlikely. The 5 months calculation would yield an increase to 250% of current network power. This is absolutely what is going to happen, in comparison.

From then on, further 20% increases are becoming more and more unlikely because of their sheer magnitude.
The more hashing power the network gets, the more stable it becomes. Right now, something like this can happen. But soon, these increases will need incredibly disruptive new technologies.

SebastianJu

legendary

Activity: 2674

Merit: 1083

Legendary Escrow Service - Tip Jar in Profile

Thanks for telling. I hope you make good use of it...

szym00

member

Activity: 74

Merit: 10

I have received my first sample chip, so thank you Sebastian. It looks that everything goes in the right direction!

svetliopi44

full member

Activity: 173

Merit: 100

Thanks for your time.
Stop to prevent spam topic the author.

svetliopi44

full member

Activity: 173

Merit: 100

I want to thank my colleague ujka . But are these 9% accurate? In my experience the difficulty increases by more than 20% of any further increase.
Please tell me where miscount, but my calculations that comes after five months would be pointless to have such a machine of 320 chips.
Please discuss, I'm new in this area and obviously wrong somewhere.
Thank you.
Sorry for bad english.

souliouz

newbie

Activity: 42

Merit: 0

Quote from: ?? on ??

https://www.dropbox.com/s/5dnr6rt4vagl0b8/90gh-august.png
added a chart, calculations done on 9% increase each retarget, earnins for a 90gh miner

Cool,
What tool or website are you using for generating this?

SebastianJu

legendary

Activity: 2674

Merit: 1083

Legendary Escrow Service - Tip Jar in Profile

Quote from: daemondazz on June 22, 2013, 04:42:29 AM

Quote from: SebastianJu on June 21, 2013, 05:49:52 AM

Strange... but at least a couple of devs got back and said they arrived. I sent out more chips today and will make the pm ready now.

The tracking number at deutsche post shows it was handed off International on the 14th. It can take 10 or so business days from Europe, so I'm not stressing just yet.

Ok... i didnt ship many items internationally so i thought it will be faster. But as long as it reaches in a good enough timeframe...

daemondazz

sr. member

Activity: 448

Merit: 250

Quote from: SebastianJu on June 21, 2013, 05:49:52 AM

Strange... but at least a couple of devs got back and said they arrived. I sent out more chips today and will make the pm ready now.

The tracking number at deutsche post shows it was handed off International on the 14th. It can take 10 or so business days from Europe, so I'm not stressing just yet.

svetliopi44

full member

Activity: 173

Merit: 100

I want to ask this increasing difficulty does it make sense to make a machine of 320 chips, which would set out to work after 5 months. What would be the difficulty and profitability in this case. For the last few days (weeks) we see a very sharp increase in the difficulties, perhaps the reason is supply machines from major manufacturers such as ASIC. For the last two weeks difficulties has risen from a 15339258.2724 19339258.2724 and after each block it up more than the previous. If difficulties continue to increase at this rate I think that after 7-8 months Machine 320 chip will not be able to repay the investment. I hope to have other colleagues forum who have surveillance on this and be willing to comment on it.

marto74

hero member

Activity: 728

Merit: 500

Thank you very much.
This will help us.

SebastianJu

legendary

Activity: 2674

Merit: 1083

Legendary Escrow Service - Tip Jar in Profile

Quote from: marto74 on June 21, 2013, 07:14:21 AM

Hi,
When you asked me a month ago if I need samples I reply atleast 2
having in mind , that I'm going to get 6 from burnin.
@ the moment We are testing both our design and also K1 helping BKKcoins.
If it's possible I'll need 4-5 more chips

Thank you in advance

Martin

You only wrote that 2 chips are ok that time. But i think its ok to send you 5 more. Its the best use for sample chips anyway.

Topic: [Group Buy] Avalon ASIC Chips (SebastianJu) Batch 6 ordered (Closed) - page 37. (Read 125282 times)