Pages:
Author

Topic: BFL: Chips have shipped, on their way to US - page 13. (Read 25583 times)

legendary
Activity: 1876
Merit: 1003
February 11, 2013, 05:19:52 PM
#32

So if i bought one today, I am waiting untill November? bull shit... they better drop that price by $300 if they think I am waiting 7 months for delivery of the products.

MFG costs should go down as time continues, or the Machines should have considerably more hashing power, or Power efficiency a year later.
If you start off with a tiny 65nm process that is extremely densely packed...

You can't expect that a 45nm or 32nm product to cost the same or less. It has to cost more. The only way to bring down the price is at the cost of profit.

http://en.wikipedia.org/wiki/Die_shrink

Die shrinks are popular among semiconductor companies, such as Intel, AMD (including the former ATI), NVIDIA, and Samsung for enriching their product lines. Examples in the 2000s include the codenamed Cedar Mill Pentium 4 processors (from 90 nm CMOS to 65 nm CMOS) and Penryn Core 2 processors (from 65 nm CMOS to 45 nm CMOS), the codenamed Brisbane Athlon 64 X2 processors (from 90 nm SOI to 65 nm SOI), and various generations of GPUs from both ATI and NVIDIA. In January 2010, Intel released Clarkdale Core i5 and Core i7 processors fabricated with a 32 nm process, down from a previous 45 nm process used in older iterations of the Nehalem processor microarchitecture.
Die shrinks are beneficial to end-users as shrinking a die reduces the current used by each transistor switching on or off in semiconductor devices while maintaining the same clock frequency of a chip, making a product with less power consumption (and thus less heat production), increased clock rate headroom, and lower prices.
The lower prices are due to mass volume production of other parts. Keep in mind BFL and Avalon are selling entire units. NOT single chips like those you cited.

They (AMD, Intel etc) also sell hundreds of millions of units (chips) so the NRE costs are a tiny fraction of the total cost. Scale <--- makes the price come down.

With ASIC's that is not the case. It is one batch for a few hundred people and only a number of wafers. (Up to a [low] multiple of 10,000 chips.)

(Note: For example, BFL only did 6 wafers for around ~6000~ ASIC chips in their first batch)

The cases and the rest of the components do not come down in price unless they establish a long term contract for the same parts. (in other words, not a single run)

This is basic economics of scale. It should be extremely obvious.

Interesting change of subject.  You were speaking about nm scale production costs.
You said "You can't expect that a 45nm or 32nm product to cost the same or less. It has to cost more. "
Which is incorrect.  It should be extremely obvious.
Wow,

So then answer this: Why didn't BFL go for 45nm technology from the start? Any guesses?

Edit: I think I understand your confusion. You think I am talking about the ASIC chips only and not the entire unit.

NO, I am referring to the entire unit. No ASIC company to date sells their chips like an AMD or Intel. Adding NRE costs to a smaller nm does increase the number of chips. But if done wrong usually results in less viable chips due to tiny defects.

AMD or Intel use binning and scale to get around that issue. As far as I know, I dont' see BFL or Avalon using binning techniques.
legendary
Activity: 1876
Merit: 1003
February 11, 2013, 05:18:39 PM
#31
Consider putting up for sale a 45nm ASIC. (Unit/powersupply/ASIC chips and all misc components.)

You are going to pay the design cost AND the mask etc.

Do you think it will cost you less or more to do business if you are only selling 5,000 units to a small community?

---------------------------

Now do the same only with 30,000,000 units. The price goes down, not up.

Everyone down the line has a contract with you in the latter scenario. Your costs go down as you buy parts in [true] bulk. Your quoted material costs per unit go down significantly compared to a tiny order.
member
Activity: 72
Merit: 10
February 11, 2013, 05:15:53 PM
#30

So if i bought one today, I am waiting untill November? bull shit... they better drop that price by $300 if they think I am waiting 7 months for delivery of the products.

MFG costs should go down as time continues, or the Machines should have considerably more hashing power, or Power efficiency a year later.
If you start off with a tiny 65nm process that is extremely densely packed...

You can't expect that a 45nm or 32nm product to cost the same or less. It has to cost more. The only way to bring down the price is at the cost of profit.

http://en.wikipedia.org/wiki/Die_shrink

Die shrinks are popular among semiconductor companies, such as Intel, AMD (including the former ATI), NVIDIA, and Samsung for enriching their product lines. Examples in the 2000s include the codenamed Cedar Mill Pentium 4 processors (from 90 nm CMOS to 65 nm CMOS) and Penryn Core 2 processors (from 65 nm CMOS to 45 nm CMOS), the codenamed Brisbane Athlon 64 X2 processors (from 90 nm SOI to 65 nm SOI), and various generations of GPUs from both ATI and NVIDIA. In January 2010, Intel released Clarkdale Core i5 and Core i7 processors fabricated with a 32 nm process, down from a previous 45 nm process used in older iterations of the Nehalem processor microarchitecture.
Die shrinks are beneficial to end-users as shrinking a die reduces the current used by each transistor switching on or off in semiconductor devices while maintaining the same clock frequency of a chip, making a product with less power consumption (and thus less heat production), increased clock rate headroom, and lower prices.
The lower prices are due to mass volume production of other parts. Keep in mind BFL and Avalon are selling entire units. NOT single chips like those you cited.

They (AMD, Intel etc) also sell hundreds of millions of units (chips) so the NRE costs are a tiny fraction of the total cost. Scale <--- makes the price come down.

With ASIC's that is not the case. It is one batch for a few hundred people and only a number of wafers. (Up to a [low] multiple of 10,000 chips.)

(Note: For example, BFL only did 6 wafers for around ~6000~ ASIC chips in their first batch)

The cases and the rest of the components do not come down in price unless they establish a long term contract for the same parts. (in other words, not a single run)

This is basic economics of scale. It should be extremely obvious.

Interesting change of subject.  You were speaking about nm scale production costs.
You said "You can't expect that a 45nm or 32nm product to cost the same or less. It has to cost more. "
Which is incorrect.  It should be extremely obvious.
legendary
Activity: 1876
Merit: 1003
February 11, 2013, 05:09:29 PM
#29

So if i bought one today, I am waiting untill November? bull shit... they better drop that price by $300 if they think I am waiting 7 months for delivery of the products.

MFG costs should go down as time continues, or the Machines should have considerably more hashing power, or Power efficiency a year later.
If you start off with a tiny 65nm process that is extremely densely packed...

You can't expect that a 45nm or 32nm product to cost the same or less. It has to cost more. The only way to bring down the price is at the cost of profit.

http://en.wikipedia.org/wiki/Die_shrink

Die shrinks are popular among semiconductor companies, such as Intel, AMD (including the former ATI), NVIDIA, and Samsung for enriching their product lines. Examples in the 2000s include the codenamed Cedar Mill Pentium 4 processors (from 90 nm CMOS to 65 nm CMOS) and Penryn Core 2 processors (from 65 nm CMOS to 45 nm CMOS), the codenamed Brisbane Athlon 64 X2 processors (from 90 nm SOI to 65 nm SOI), and various generations of GPUs from both ATI and NVIDIA. In January 2010, Intel released Clarkdale Core i5 and Core i7 processors fabricated with a 32 nm process, down from a previous 45 nm process used in older iterations of the Nehalem processor microarchitecture.
Die shrinks are beneficial to end-users as shrinking a die reduces the current used by each transistor switching on or off in semiconductor devices while maintaining the same clock frequency of a chip, making a product with less power consumption (and thus less heat production), increased clock rate headroom, and lower prices.
The lower prices are due to mass volume production of other parts. Keep in mind BFL and Avalon are selling entire units. NOT single chips like those you cited.

They (AMD, Intel etc) also sell hundreds of millions of units (chips) so the NRE costs are a tiny fraction of the total cost. Scale <--- makes the price come down.

With ASIC's that is not the case. It is one batch for a few hundred people and only a number of wafers. (Up to a [low] multiple of 10,000 chips.)

(Note: For example, BFL only did 6 wafers for around ~6000~ ASIC chips in their first batch)

The cases and the rest of the components do not come down in price unless they establish a long term contract for the same parts. (in other words, not a single run)

This is basic economics of scale. It should be extremely obvious.
hero member
Activity: 868
Merit: 1000
February 11, 2013, 05:05:03 PM
#28
Have BFL told people to start sending back their FPGAs yet?  They said they would do this at least 7 days before shipping ASICs so it needs to happen in the next few days if they're planning to ship on 22 February.
legendary
Activity: 1274
Merit: 1004
February 11, 2013, 04:40:34 PM
#27

So if i bought one today, I am waiting untill November? bull shit... they better drop that price by $300 if they think I am waiting 7 months for delivery of the products.

MFG costs should go down as time continues, or the Machines should have considerably more hashing power, or Power efficiency a year later.
If you start off with a tiny 65nm process that is extremely densely packed...

You can't expect that a 45nm or 32nm product to cost the same or less. It has to cost more. The only way to bring down the price is at the cost of profit.

http://en.wikipedia.org/wiki/Die_shrink

Die shrinks are popular among semiconductor companies, such as Intel, AMD (including the former ATI), NVIDIA, and Samsung for enriching their product lines. Examples in the 2000s include the codenamed Cedar Mill Pentium 4 processors (from 90 nm CMOS to 65 nm CMOS) and Penryn Core 2 processors (from 65 nm CMOS to 45 nm CMOS), the codenamed Brisbane Athlon 64 X2 processors (from 90 nm SOI to 65 nm SOI), and various generations of GPUs from both ATI and NVIDIA. In January 2010, Intel released Clarkdale Core i5 and Core i7 processors fabricated with a 32 nm process, down from a previous 45 nm process used in older iterations of the Nehalem processor microarchitecture.
Die shrinks are beneficial to end-users as shrinking a die reduces the current used by each transistor switching on or off in semiconductor devices while maintaining the same clock frequency of a chip, making a product with less power consumption (and thus less heat production), increased clock rate headroom, and lower prices.

That's only true because you can fit more of the smaller dies on the same wafer, even though the wafer itself costs more. That doesn't take into account the NRE involved with the project though. In the case where volume is small like it is for BTC ASICs, you might never save enough per die to offset the cost of all the new tooling, let alone pass the savings on to your customers.
member
Activity: 72
Merit: 10
February 11, 2013, 04:32:26 PM
#26

So if i bought one today, I am waiting untill November? bull shit... they better drop that price by $300 if they think I am waiting 7 months for delivery of the products.

MFG costs should go down as time continues, or the Machines should have considerably more hashing power, or Power efficiency a year later.
If you start off with a tiny 65nm process that is extremely densely packed...

You can't expect that a 45nm or 32nm product to cost the same or less. It has to cost more. The only way to bring down the price is at the cost of profit.

http://en.wikipedia.org/wiki/Die_shrink

Die shrinks are popular among semiconductor companies, such as Intel, AMD (including the former ATI), NVIDIA, and Samsung for enriching their product lines. Examples in the 2000s include the codenamed Cedar Mill Pentium 4 processors (from 90 nm CMOS to 65 nm CMOS) and Penryn Core 2 processors (from 65 nm CMOS to 45 nm CMOS), the codenamed Brisbane Athlon 64 X2 processors (from 90 nm SOI to 65 nm SOI), and various generations of GPUs from both ATI and NVIDIA. In January 2010, Intel released Clarkdale Core i5 and Core i7 processors fabricated with a 32 nm process, down from a previous 45 nm process used in older iterations of the Nehalem processor microarchitecture.
Die shrinks are beneficial to end-users as shrinking a die reduces the current used by each transistor switching on or off in semiconductor devices while maintaining the same clock frequency of a chip, making a product with less power consumption (and thus less heat production), increased clock rate headroom, and lower prices.
legendary
Activity: 1876
Merit: 1003
February 11, 2013, 04:27:47 PM
#25

So if i bought one today, I am waiting untill November? bull shit... they better drop that price by $300 if they think I am waiting 7 months for delivery of the products.

MFG costs should go down as time continues, or the Machines should have considerably more hashing power, or Power efficiency a year later.
If you start off with a tiny 65nm process that is extremely densely packed...

You can't expect that a 45nm or 32nm product to cost the same or less. It has to cost more. The only way to bring down the price is at the cost of profit.
newbie
Activity: 56
Merit: 0
February 10, 2013, 03:36:00 PM
#24
Josh has stated today that the Fedex tracking # shows it arriving at the bumping facility in California on Thursday.  Personally I think they start shipping the final week of February, but we'll see.

hero member
Activity: 742
Merit: 500
February 10, 2013, 02:24:44 PM
#23
Pringles.png

Don't pop your top just yet.  They still need to be bumped, packaged, and most importantly, thoroughly tested.

This ^^

Some of you are blind aren't you? There hasn't been really any tests on the BFL "chips" that are "on the way to the US".

Hurry up and WAIT lol

Suckerz....  Grin Grin Grin
I have to agree. I just posted a thread about this (that was deleted for some reason). These chips still have to make their way across the world to the US (i'm willing to bet they're not overnight airmailed....), assembled, and tested. That process alone will take weeks, and much longer if there's any sort of issues (let's face it, this is untested product - there will be issues). I was very late to the party (order number 18k), and i honestly don't expect to receive mine until well after the summer. I won't be able to get in early, or recoup my investment quickly, but let's face it - a jalapeno is going to be the minimum you'll need to make any sort of profit mining here very soon.
legendary
Activity: 889
Merit: 1000
Bitcoin calls me an Orphan
February 10, 2013, 12:34:02 PM
#22
Nice to see the chips where shipped in bulk. I hope they have more chips on order since they have so many orders in place.
sr. member
Activity: 428
Merit: 250
February 10, 2013, 12:19:27 PM
#21
It would be GOD DAMN HILARIOUS if BFL have a dozen confirmed received ASICs and AVALON still only have 2.5 in the wild after "shipping"

Tongue

No what would be funnier that BFL with all their claims of chips being shipped to the U.S. found out that they never got shipped and they now have another 30 delay on the chips being shipped. LOL seeing you fools with your heads hung after 7-8 months of waiting. Fucking priceless.



I'm also waiting to see what's going to happend with BFL in next 15 days as I was sure they won't deliver in february even in March... Even if I hope for their customer they do respect finally theor word after 10 delay story...
legendary
Activity: 2492
Merit: 1473
LEALANA Bitcoin Grim Reaper
February 10, 2013, 06:03:22 AM
#20
It would be GOD DAMN HILARIOUS if BFL have a dozen confirmed received ASICs and AVALON still only have 2.5 in the wild after "shipping"

Tongue

No what would be funnier that BFL with all their claims of chips being shipped to the U.S. found out that they never got shipped and they now have another 30 delay on the chips being shipped. LOL seeing you fools with your heads hung after 7-8 months of waiting. Fucking priceless.

hero member
Activity: 896
Merit: 532
Former curator of The Bitcoin Museum
February 10, 2013, 05:36:24 AM
#19
It would be GOD DAMN HILARIOUS if BFL have a dozen confirmed received ASICs and AVALON still only have 2.5 in the wild after "shipping"

Tongue
full member
Activity: 196
Merit: 100
February 10, 2013, 04:22:37 AM
#18
Gosh, I ordered in June as well so I better be in this first batch!  I'm really hoping it was worth the wait.

What do you have pre-ordered?
hero member
Activity: 608
Merit: 500
February 10, 2013, 02:58:28 AM
#17
Gosh, I ordered in June as well so I better be in this first batch!  I'm really hoping it was worth the wait.
legendary
Activity: 2492
Merit: 1473
LEALANA Bitcoin Grim Reaper
February 10, 2013, 12:08:51 AM
#16
Pringles.png

Don't pop your top just yet.  They still need to be bumped, packaged, and most importantly, thoroughly tested.

This ^^

Some of you are blind aren't you? There hasn't been really any tests on the BFL "chips" that are "on the way to the US".

Hurry up and WAIT lol

Suckerz....  Grin Grin Grin
full member
Activity: 196
Merit: 100
February 09, 2013, 09:47:48 PM
#15
Be prepared for CNY surprises soon:) hot news are knocking on the door

Yeah, it sucks how the whole US practically shuts down at CNY.
legendary
Activity: 952
Merit: 1000
February 09, 2013, 06:08:35 PM
#14
Pringles.png
Don't pop your top just yet.  They still need to be bumped, packaged, and most importantly, thoroughly tested.
Oh I realize that. I just like the pun that not all chip bags are all air.
legendary
Activity: 966
Merit: 1000
February 09, 2013, 05:34:32 PM
#13
Pringles.png

Don't pop your top just yet.  They still need to be bumped, packaged, and most importantly, thoroughly tested.
Pages:
Jump to: