@zum
I have PM'd Ken with my three questions
Topic: [ActiveMining] The Official Active Mining Discussion Thread [Self-Moderated] - page 80. (Read 771280 times)
January 26, 2014, 07:12:49 PM
January 26, 2014, 07:07:48 PM
Those numbers seem way off. For example, Bitfury chips get around 2GH/s and they probably get ~4k dies out of a single wafer. Your cost for a fully processed wafer is also low, though not enough to even out with being possibly 20x on the number of chips.
Cost should be a lot lower than $2/GH/s at the board level to have chance of being competitive though. Cointerra is selling May delivery units in a chassis with cooling, controller and power supplies for $3/GH/s, and that's including markup to cover overhead. Cost for a 55nm product with chips shipping even in Q2 will have to be way below $1GH/s at the board level to be sensible.
Cost should be a lot lower than $2/GH/s at the board level to have chance of being competitive though. Cointerra is selling May delivery units in a chassis with cooling, controller and power supplies for $3/GH/s, and that's including markup to cover overhead. Cost for a 55nm product with chips shipping even in Q2 will have to be way below $1GH/s at the board level to be sensible.
Right, should have thought longer about that. I don't have any access to the latest reports behind paywalls, or needing industry membership, I was seeing 2 year old projections showing 55nm dropping below $1000 a wafer and figuring UMC being the cheapest in the business.
Also, yah, should have thought twice about number per wafer, think it was GPU cores, which can be massive, BFL got 1000 per wafer didn't they on 65, for 4gh chips so don't know what I was thinking there.
January 26, 2014, 07:00:48 PM
@minerpart
The question is for Ken to answer. One would hope he knows the costs...
The question is for Ken to answer. One would hope he knows the costs...
Talking of which, Ken got back to me to say he would address the questions I posted earlier in the weekly update. I have taken no action regarding your subsequent questions.
January 26, 2014, 07:00:06 PM
Bloody 'el, calm your jets!
I have re-read the last couple of pages and agree I was mistaken to assume that Vigil was having a go. He was saying what he believed to be factual and if I'd not missed the crucial coma in his sentence perhaps I wouldn't have jumped to the wrong conclusion. My judgement was clouded by Vigil's recent turn from pro-ActM to calling it a scam and having cheap digs at such things like the volume on CT after Vigils theory that we were already mining was shown to be wrong.
I apologise sincerely for making this rash judgement, for the unfortunate language I used and for upsetting anyone who I may have offended.
Your turn..
How is citing CT's volume a cheap dig? Also, my theory was never actually shown to be wrong. People just claimed it was wrong. We don't have enough information to prove anything. Its still a possibility in my book, regardless, until the owner of that operation steps forward and says otherwise. Did I ever acknowledge that it was incorrect? I have re-read the last couple of pages and agree I was mistaken to assume that Vigil was having a go. He was saying what he believed to be factual and if I'd not missed the crucial coma in his sentence perhaps I wouldn't have jumped to the wrong conclusion. My judgement was clouded by Vigil's recent turn from pro-ActM to calling it a scam and having cheap digs at such things like the volume on CT after Vigils theory that we were already mining was shown to be wrong.
I apologise sincerely for making this rash judgement, for the unfortunate language I used and for upsetting anyone who I may have offended.
Your turn..
January 26, 2014, 06:50:33 PM
@minerpart
The question is for Ken to answer. One would hope he knows the costs...
The question is for Ken to answer. One would hope he knows the costs...
January 26, 2014, 06:43:06 PM
For just the 55nm chips, I'm spitballing numbers that look something like, $500 a wafer, 200 chips a wafer, 2gh a chip, 10c a chip for balling and packaging. So about a buck thirty a gigahash in chips, say a buck 50 with NRE amortized over a crap load. Probably $2 a Gh our cost by the time they are on boards...
If the numbers come close to that ballpark, it's making sense.
Those numbers seem way off. For example, Bitfury chips get around 2GH/s and they probably get ~4k dies out of a single wafer. Your cost for a fully processed wafer is also low, though not enough to even out with being possibly 20x on the number of chips.If the numbers come close to that ballpark, it's making sense.
Cost should be a lot lower than $2/GH/s at the board level to have chance of being competitive though. Cointerra is selling May delivery units in a chassis with cooling, controller and power supplies for $3/GH/s, and that's including markup to cover overhead. Cost for a 55nm product with chips shipping even in Q2 will have to be way below $1GH/s at the board level to be sensible.
January 26, 2014, 06:40:39 PM
"UMC first taped-out customer 55nm SDDI products in late 2012. Achieving 15 million accumulated shipments and mature production yields in just one year is a significant milestone and highlights our engineering and manufacturing strength as a worldwide foundry specialty technology leader. With our 300mm Fab 12A in Tainan, Taiwan and Singapore's 300mm Fab 12i both providing ample capacity support and established economies of scale, we look forward to bringing additional 55nm products to volume production as many new products are scheduled for the design-in stage in early 2014."
http://finance.yahoo.com/news/umc-surpasses-15-million-shipments-100000385.html
http://finance.yahoo.com/news/umc-surpasses-15-million-shipments-100000385.html
January 26, 2014, 06:25:06 PM
2) What were the fixed costs for the 55nm chip. This includes (but not limited to) IP costs from the two new engineers (monetary or otherwise), NRE costs for the tapeout, fixed design costs for boards (if not amortised into the per unit deployment cost in Q1).
Well the chip is not yet in production so the costs are not yet known.
We have taped-out. We don't know when that was done and we don't know how close we are to the 'Silicon Process'. So the cost of getting the chip working on a board is yet to be discovered.
http://www.umc.com/english/design/e.asp
January 26, 2014, 06:20:58 PM
For just the 55nm chips, I'm spitballing numbers that look something like, $500 a wafer, 200 chips a wafer, 2gh a chip, 10c a chip for balling and packaging. So about a buck thirty a gigahash in chips, say a buck 50 with NRE amortized over a crap load. Probably $2 a Gh our cost by the time they are on boards...
If the numbers come close to that ballpark, it's making sense.
If the numbers come close to that ballpark, it's making sense.
January 26, 2014, 05:25:10 PM
@ Vince and Zumzero
For the weekly Q&A
1) What is the estimated $ per GH costs for the full deployment of the 55nm chips. This includes (but not limited to) chip cost, board cost, controller cost, set up costs, coding costs, rent, power, staffing, security etc.
2) What were the fixed costs for the 55nm chip. This includes (but not limited to) IP costs from the two new engineers (monetary or otherwise), NRE costs for the tapeout, fixed design costs for boards (if not amortised into the per unit deployment cost in Q1).
3) What are the limits (if any) to the space/power available in our chosen data centre.
Those are my top three questions. I don't want to add any more in case these three are not answered. I hope they are not brushed aside.
For the weekly Q&A
1) What is the estimated $ per GH costs for the full deployment of the 55nm chips. This includes (but not limited to) chip cost, board cost, controller cost, set up costs, coding costs, rent, power, staffing, security etc.
2) What were the fixed costs for the 55nm chip. This includes (but not limited to) IP costs from the two new engineers (monetary or otherwise), NRE costs for the tapeout, fixed design costs for boards (if not amortised into the per unit deployment cost in Q1).
3) What are the limits (if any) to the space/power available in our chosen data centre.
Those are my top three questions. I don't want to add any more in case these three are not answered. I hope they are not brushed aside.
January 26, 2014, 04:49:21 PM
At this specific point in time, I don't know why anyone would be expecting anything more than another delay.
January 26, 2014, 04:22:33 PM
Oh, whoosh.
Anyways guys!
The 7th Day is onTuesday (I guess Thursday? is 7 Days after the 22nd) I believe.
Anyways guys!
We are selling Ukyo shares on Crypto-Trade to recover our losses on Bitfunder.
We will start moving Investors shares to Crypto-Trade within the next 7 days.
We will start moving Investors shares to Crypto-Trade within the next 7 days.
The 7th Day is on
the 29th will be exactly 7 days from the 22nd. My guess is it will start in earnest tomorrow.
January 26, 2014, 04:00:41 PM
Oh, whoosh.
Anyways guys!
The 7th Day is onTuesday (I guess Thursday? is 7 Days after the 22nd) I believe.
Anyways guys!
We are selling Ukyo shares on Crypto-Trade to recover our losses on Bitfunder.
We will start moving Investors shares to Crypto-Trade within the next 7 days.
We will start moving Investors shares to Crypto-Trade within the next 7 days.
The 7th Day is on
January 26, 2014, 03:45:55 PM
HOW THE F#@K IS THIS GROSSLY UNFAIR
First off calm down, do you really need to get so excited?
Secondly Ken has done all the calulations to see that commisioning the 55nm chip will make money. The question was about the price-point in the market that the chip would take. Ken was saying he has yet to price the chip for the retail and bulk-buy market.
I know it is hard for a lot of you to keep up with what's going on but please try to stay calm if you don't understand something instead of jumping to conclusions and having a shouting match.
Haven't you heard? This is how you get the privilege of the title, "Realist" around here:
1. Take a quote (feel free to disregard context!)
2. Make an assumption on said quote (ensure that whatever liberties you take with the assumption fit your mood and/or agenda)
3. RAAAAAAAAAAAAAAGGGGGEEEEEEEEEEE!
4. Pretend your assumptions are now facts but don't you dare feel the need to prove this! <-- VERY IMPORTANT STEP!
5. Label all counter-arguments as "cheer leading".
6. Pretend that your method of deduction is logical. <-- Optional step. Feel free to skip.
7. Rage some more. Harder this time. I SAID HARDER!
8. More bitching about cheer leading, disregard arguments that you whole argument is based on an assumption that is probably wrong. It doesn't matter. You're MAD!
9. Enter circular reasoning.
10. Enjoy being the realist around here! God knows we need you!
Step 7 is definitely optional too because I think I am a realist. I don't remember raging.
None of these steps apply to a true realist. That's the joke!
January 26, 2014, 03:33:33 PM
Well just as well I PM'd Ken on your behalf.
You've PM'd him with all the questions that have been ignored/unanswered?
gee - must have been a long PM
No, I asked three questions related to the points you have raised in the last couple of hours. I hope you don't mind me taking the liberty of adding my own twist of diplomacy as I know it can be incongruous to your methods.
Actually you're right - I guess i need to reassess my patience levels again. Hope you get some decent answers - mind posting the exact questions here while we wait for the big K-dawg to answer? I'd like to see the wording
My pleasure. Although I think if Vince is happy to continue with organising the weekly Q&A he should be our point of contact.
Hi Ken,
Following some recent activity on the thread I thought I'd take it upon myself to put a couple of things to you. If you have the time and are happy to respond then I would be looking to post your reply on the thread. If you are unable to do so I respect that and will wait for the weekly update.
Without committing to a specific date are you able to estimate how
long until the 55nm chips are ready?
Are you able to provide any details which support the decision to pursue the development of the 55nm chips understanding that it is popular belief that this technology is approaching the end of its useful lifespan in Bitcoin mining.
Will the boards and controllers for the 55nm chips be the same as the 28nm chips?
Cheers,
zumzero
Following some recent activity on the thread I thought I'd take it upon myself to put a couple of things to you. If you have the time and are happy to respond then I would be looking to post your reply on the thread. If you are unable to do so I respect that and will wait for the weekly update.
Without committing to a specific date are you able to estimate how
long until the 55nm chips are ready?
Are you able to provide any details which support the decision to pursue the development of the 55nm chips understanding that it is popular belief that this technology is approaching the end of its useful lifespan in Bitcoin mining.
Will the boards and controllers for the 55nm chips be the same as the 28nm chips?
Cheers,
zumzero
January 26, 2014, 03:23:59 PM
HOW THE F#@K IS THIS GROSSLY UNFAIR
First off calm down, do you really need to get so excited?
Secondly Ken has done all the calulations to see that commisioning the 55nm chip will make money. The question was about the price-point in the market that the chip would take. Ken was saying he has yet to price the chip for the retail and bulk-buy market.
I know it is hard for a lot of you to keep up with what's going on but please try to stay calm if you don't understand something instead of jumping to conclusions and having a shouting match.
Haven't you heard? This is how you get the privilege of the title, "Realist" around here:
1. Take a quote (feel free to disregard context!)
2. Make an assumption on said quote (ensure that whatever liberties you take with the assumption fit your mood and/or agenda)
3. RAAAAAAAAAAAAAAGGGGGEEEEEEEEEEE!
4. Pretend your assumptions are now facts but don't you dare feel the need to prove this! <-- VERY IMPORTANT STEP!
5. Label all counter-arguments as "cheer leading".
6. Pretend that your method of deduction is logical. <-- Optional step. Feel free to skip.
7. Rage some more. Harder this time. I SAID HARDER!
8. More bitching about cheer leading, disregard arguments that you whole argument is based on an assumption that is probably wrong. It doesn't matter. You're MAD!
9. Enter circular reasoning.
10. Enjoy being the realist around here! God knows we need you!
Step 7 is definitely optional too because I think I am a realist. I don't remember raging.
January 26, 2014, 03:23:52 PM
Well just as well I PM'd Ken on your behalf.
You've PM'd him with all the questions that have been ignored/unanswered?
gee - must have been a long PM
No, I asked three questions related to the points you have raised in the last couple of hours. I hope you don't mind me taking the liberty of adding my own twist of diplomacy as I know it can be incongruous to your methods.
Actually you're right - I guess i need to reassess my patience levels again. Hope you get some decent answers - mind posting the exact questions here while we wait for the big K-dawg to answer? I'd like to see the wording
January 26, 2014, 03:04:31 PM
wow biggest bitcoin scam ever lol unlucky
January 26, 2014, 02:51:59 PM
Quote
HOW THE FUCK IS THIS GROSSLY UNFAIR AND A CHEAP DIG??? KEN ADMITTED IT AFTER I POSTED A BRIEF RUNDOWN OF OTHER 55nm CHIPS?!?!? SERIOUSLY WHAT IS WRONG WITH YOU?
Bloody 'el, calm your jets!
I have re-read the last couple of pages and agree I was mistaken to assume that Vigil was having a go. He was saying what he believed to be factual and if I'd not missed the crucial coma in his sentence perhaps I wouldn't have jumped to the wrong conclusion. My judgement was clouded by Vigil's recent turn from pro-ActM to calling it a scam and having cheap digs at such things like the volume on CT after Vigils theory that we were already mining was shown to be wrong.
I apologise sincerely for making this rash judgement, for the unfortunate language I used and for upsetting anyone who I may have offended.
Your turn..
January 26, 2014, 02:43:31 PM
HOW THE F#@K IS THIS GROSSLY UNFAIR
First off calm down, do you really need to get so excited?
Secondly Ken has done all the calulations to see that commisioning the 55nm chip will make money. The question was about the price-point in the market that the chip would take. Ken was saying he has yet to price the chip for the retail and bulk-buy market.
I know it is hard for a lot of you to keep up with what's going on but please try to stay calm if you don't understand something instead of jumping to conclusions and having a shouting match.
Haven't you heard? This is how you get the privilege of the title, "Realist" around here:
1. Take a quote (feel free to disregard context!)
2. Make an assumption on said quote (ensure that whatever liberties you take with the assumption fit your mood and/or agenda)
3. RAAAAAAAAAAAAAAGGGGGEEEEEEEEEEE!
4. Pretend your assumptions are now facts but don't you dare feel the need to prove this! <-- VERY IMPORTANT STEP!
5. Label all counter-arguments as "cheer leading".
6. Pretend that your method of deduction is logical. <-- Optional step. Feel free to skip.
7. Rage some more. Harder this time. I SAID HARDER!
8. More bitching about cheer leading, disregard arguments that you whole argument is based on an assumption that is probably wrong. It doesn't matter. You're MAD!
9. Enter circular reasoning.
10. Enjoy being the realist around here! God knows we need you!
Jump to: