GenTarkin
Legendary
Offline
Activity: 2450
Merit: 1002
|
|
February 06, 2015, 03:14:05 AM |
|
Edit: What they said ^ I think the above was just Spondoolies-Tech's slightly unsubtly way of saying: Hey, our software is open source. Here's our git repository. Feel free to fork it, make changes - such as the suggested Nelder–Mead approach for N-dimensional optimization, and submit pull requests, and we'd be happy to review the changes and possible integrate them into our baseline - or you can provide your own custom builds for users who would be interested in it. Code repositories are generally not intended for end-users My not so subtle response to this: Spondoolies hired a single junior/young software developer who had landed far out of his depth with regards to state-of-the-art of the optimization of rather complex analog/digital system. His attempts at one-dimensional optimization problems when searching for the optimal configuration parameters show that he isn't aware of http://en.wikipedia.org/wiki/Golden_section_search known since mid-1950 or mid-1960. And various folks here already started to request optimization in 2 or more dimensions. This is going to end up rather badly with regards to the software quality when the corporation gives his lead developer no time to actually research the problem space. Instead they gave him a mishmash of tasks from the request list that are rather trivial and cosmetic. So my friendly advice to zvisha is: step away from fulfilling trivial maintenance requests and broaden your knowledge with a quick peek into gradient-free/derivative-free optimization methods that I've mentioned. The payoff to him (personally) and the company and its user base will be much higher. My not so subtle advice to Guy: you'll need to give time to your staff to do an actual research as opposed to bogging them down with constant development and maintenance requests. Your company's current work on finding the optimal operating point for your miners is equivalent to the state-of-the-art around World War II. Either give zvisha some time off of D and let him concentrate on R (from R&D) or hire a actual mathematical optimization consultant. This should be quite easy in your location in Israel. I actually don't care if you open source it or keep it as a closed-source module that controls the cgminer through its API. You could probably then charge extra for it or differentiate your product from the competition in marketing. The above was a free advice. If you don't like it just ask and I will give you a full refund of the money paid. I agree, just watching how the miner behaves when I specify a working max Wattage, my units are clearly not getting the best GH/W possible. Like I said, they clock themselves down for no apparent reason... Also, an error rate of 2% is acceptable, these things allow virtually no errors, which I believe is the reason for the aggressive declocking =( I bet I could hand tune clocks / voltages statically better than their autotuning does...
|
|
|
|
RoadStress
Legendary
Offline
Activity: 1904
Merit: 1007
|
|
February 06, 2015, 03:17:12 AM |
|
I agree, just watching how the miner behaves when I specify a working max Wattage, my units are clearly not getting the best GH/W possible. Like I said, they clock themselves down for no apparent reason... Also, an error rate of 2% is acceptable, these things allow virtually no errors, which I believe is the reason for the aggressive declocking =(
I bet I could hand tune clocks / voltages statically better than their autotuning does...
But is it worth it? How much would you gain? If it's less than 5% then it's not such a big gain...
|
|
|
|
2112
Legendary
Offline
Activity: 2128
Merit: 1073
|
|
February 06, 2015, 03:24:52 AM |
|
But is it worth it? How much would you gain? If it's less than 5% then it's not such a big gain...
"Worth it" to whom? To GenTrakin personally or to the Spondoolies' Corporation and their future industrial scale mining operation partners? 5% may be nothing to a piker but the same 5% yield improvement at the industrial scale means megabucks.
|
|
|
|
GenTarkin
Legendary
Offline
Activity: 2450
Merit: 1002
|
|
February 06, 2015, 03:26:37 AM |
|
But is it worth it? How much would you gain? If it's less than 5% then it's not such a big gain...
"Worth it" to whom? To GenTrakin personally or to the Spondoolies' Corporation and their future industrial scale mining operation partners? 5% may be nothing to a piker but the same 5% yield improvement at the industrial scale means megabucks. It would be worth it to me, 5% improvement in efficiency when its hard enough barely breaking even... is all the difference.
|
|
|
|
2112
Legendary
Offline
Activity: 2128
Merit: 1073
|
|
February 06, 2015, 03:34:29 AM |
|
It would be worth it to me, 5% improvement in efficiency when its hard enough barely breaking even... is all the difference.
I don't want to disagree with you. I just want to give you a different perspective. You probably discounting your per hour work time charge to some very low rate. I presume that the industrial-scale miners will not be able to hire highly conscientious people like you to run their operations. Their employees will be more like Bruce Peterson of the BFL's burn-in-room fame.
|
|
|
|
Guy Corem (OP)
Donator
Legendary
Offline
Activity: 1414
Merit: 1051
Spondoolies, Beam & DAGlabs
|
|
February 06, 2015, 04:28:48 AM |
|
Edit: What they said ^ I think the above was just Spondoolies-Tech's slightly unsubtly way of saying: Hey, our software is open source. Here's our git repository. Feel free to fork it, make changes - such as the suggested Nelder–Mead approach for N-dimensional optimization, and submit pull requests, and we'd be happy to review the changes and possible integrate them into our baseline - or you can provide your own custom builds for users who would be interested in it. Code repositories are generally not intended for end-users I wish this was true. You can fork it and use it as you code the changes, but Zvi has limited time to review any pulls. I can still see that the UI on the latest SP 20 still have some errors which I tried to rectify since I got my SP30 .. forked the stuff, did the changes, which were never implemented by SP-T due to the lack of time .. The pull request is still there .. but I stopped asking, eventually in the end, the units are hashing .. Nevertheless, code repositories are good, and I'm an end-user. Probably the exception which confirms the rule. Correct.
Checking. Please email Zvisha as well about it. Thanks
|
|
|
|
Guy Corem (OP)
Donator
Legendary
Offline
Activity: 1414
Merit: 1051
Spondoolies, Beam & DAGlabs
|
|
February 06, 2015, 04:29:36 AM |
|
I have found and examined the "SP20 Jackson Quick Start Guide". At the tail end of that 9 page document there is a reference to "SP20 Jackson User Guide" for further details. I have been unable to find the "SP20 User Guide" anywhere. Does this document exist, and if so where? Is the "SP3x User Guide" useful at all for the SP20? Sorry if this is an silly question, but I couldn't find anything relevant here.
We'll release the SP20 User Guide soon, Barbara is still working on it. We're very small team.
|
|
|
|
Guy Corem (OP)
Donator
Legendary
Offline
Activity: 1414
Merit: 1051
Spondoolies, Beam & DAGlabs
|
|
February 06, 2015, 04:39:10 AM |
|
^I think you are making a trivial issue into a major product fault that it isnt.
The Sp20 is already leading-edge for efficiency, and while theres a little space to improve efficiency with a little bit finer tuning processes, it might only be by 0.02-0.04w/GH
I think that you are wrong. I'm following this thread from the beginning and observed a very rapid release cycle with multiple regressions which is indicative of the developers thrashing around without understanding what is going on with the product in the field. I also noticed multiple user complaints about things like "entire ASIC loops disabled by BIST". I believe zvisha has already discovered a while back that GO/NO GO BIST that he had inherited from the Spondoolies' hardware designer is counterproductive and overly cautious. The use of BIST (in the field) should be completely abandoned now, it could be still valuable in the factory or in the pre-final-assembly stages. I don't know what is the internal personal dynamics of Spondoolies as a corporation. Does zvisha have enough inside clout to say NO! to some rather trivial support requests via Skype/ssh/Teamviewer? Can he influence the hardware designer to give him more information from the chip that just a single go/no-go bit and a temperature? What is required now is not more of "finer tuning". There's a need to step back and do some real research and appraisal of the entire field and how it is changing to more and more unattended operation. The cute "Thank you for the personal attention!" notices from the customers may be pleasant now, but they are actually indicative of the production problems in the factory. Thank you for your detailed feedback. I'll be happy to provide the internal documentation (Sitara - FPGA interface and ASIC interface) to anyone interested in trying to squeeze more out of RockerBox based products. The only conditions is signing mutual NDA. We won't be able to support such an effort, as our team is overstretched as it is. Guy
|
|
|
|
Bicknellski
|
|
February 06, 2015, 04:59:22 AM |
|
It would be worth it to me, 5% improvement in efficiency when its hard enough barely breaking even... is all the difference.
I don't want to disagree with you. I just want to give you a different perspective. You probably discounting your per hour work time charge to some very low rate. I presume that the industrial-scale miners will not be able to hire highly conscientious people like you to run their operations. Their employees will be more like Bruce Peterson of the BFL's burn-in-room fame. Invoking Bruce's name in a Spondoolies thread... the ultimate contrast.
|
|
|
|
alh
Legendary
Offline
Activity: 1846
Merit: 1052
|
|
February 06, 2015, 05:26:23 AM |
|
I have found and examined the "SP20 Jackson Quick Start Guide". At the tail end of that 9 page document there is a reference to "SP20 Jackson User Guide" for further details. I have been unable to find the "SP20 User Guide" anywhere. Does this document exist, and if so where? Is the "SP3x User Guide" useful at all for the SP20? Sorry if this is an silly question, but I couldn't find anything relevant here.
We'll release the SP20 User Guide soon, Barbara is still working on it. We're very small team. Thanks, I appreciate the candor. I also appreciate the support offered on this forum.
|
|
|
|
zvisha
|
|
February 06, 2015, 08:33:58 AM |
|
Edit: What they said ^ I think the above was just Spondoolies-Tech's slightly unsubtly way of saying: Hey, our software is open source. Here's our git repository. Feel free to fork it, make changes - such as the suggested Nelder–Mead approach for N-dimensional optimization, and submit pull requests, and we'd be happy to review the changes and possible integrate them into our baseline - or you can provide your own custom builds for users who would be interested in it. Code repositories are generally not intended for end-users I wish this was true. You can fork it and use it as you code the changes, but Zvi has limited time to review any pulls. I can still see that the UI on the latest SP 20 still have some errors which I tried to rectify since I got my SP30 .. forked the stuff, did the changes, which were never implemented by SP-T due to the lack of time .. The pull request is still there .. but I stopped asking, eventually in the end, the units are hashing .. Nevertheless, code repositories are good, and I'm an end-user. Probably the exception which confirms the rule. Correct.
Yes, I am sorry. I also have pending pull request from Luke which is probably good, but I do not have the time to make the integration (with all the release cycle, and testing on all the miner favours). Especially in the web, which is not my strongest side, I am reluctant to change things But developers are welcome to do custom builds, and develop their own flavours if they feel it is interesting for them.
|
|
|
|
mavericklm
|
|
February 06, 2015, 12:30:11 PM Last edit: February 06, 2015, 12:43:35 PM by mavericklm |
|
Current firmware version: 2.6.12 shutdown button should have an yes and cancel question, not directly shutdown!
what can i do about this: Feb 6 11:59:20 miner local1.warn minergate[6949]: Critical: Lazy ASIC [2] - consider disabling it if this happens more then every 10 minutes
and this: LOOP[0] OFF TO:0 (test serial failed or something) 0: disabled (serial problem) 1: disabled (serial problem)
|
|
|
|
Guy Corem (OP)
Donator
Legendary
Offline
Activity: 1414
Merit: 1051
Spondoolies, Beam & DAGlabs
|
|
February 06, 2015, 01:22:57 PM |
|
We'll soon be out of stock on the SP35.
|
|
|
|
wh00per
|
|
February 06, 2015, 02:38:22 PM Last edit: February 06, 2015, 02:52:51 PM by wh00per |
|
Yes, I am sorry. I also have pending pull request from Luke which is probably good, but I do not have the time to make the integration (with all the release cycle, and testing on all the miner flavors). Especially in the web, which is not my strongest side, I am reluctant to change things But developers are welcome to do custom builds, and develop their own flavors if they feel it is interesting for them. Zvi, there's no reason to say "I'm sorry" In my opinion the code repositories should be used interactively. Some of the guys here are addicted enough to even code stuff for you (web or not), but unfortunately they can't integrate their stuff in the mainstream. That's why 2112 suggested a step back from day to day issues and look back in with a bigger perspective. You have a unique position being on the team since the first SP-T chip was being created. Since SP35 is almost out of stock, with the new chips (if any) one can try a new approach. It will be a pity to use the same (or slightly modified) chip optimization model since the SP10 units From a personal perspective I know it's hard to break routines and take new challenges when there is a multitude of small issues to be solved. However, those issues will always be there. Sometimes I'd like to code stuff instead of driving my team to do the coding Particularly when the team is small and busy. But they need guidance and planning while they do what they do best: code. They can't be efficient when they focus on both current issues and future planning. Someone needs to do the planning for them It's the "management" job security paradigm With the continuous increase in the number of SP-T models, just testing new firmware versions for all situations out there will eventually stress your team more than it can cope with. If I were you, I'd drop support for older models (I'm still happy with my SP10 running FW 1.5.8 and I'm not looking to upgrade it any longer). That will let you focus on the new ones. Also, try an "agile" way of running things. Let your guys focus on the improvements with the largest impact and with minimum work-hours .. the R&D supporting the firmware using the future line of chips should be one of those. By the way, no one needs a weekly FW update. It can be bi-weekly or even monthly. There's no time to upgrade and monitor the units (particularly in large numbers) for improvement, that quick. What it really boils out of it, is to sort out priorities, communicating the sort result to the team. And one can't do that properly when (s)he's always busy with small stuff.
|
CSA/cUL Certified Power Distribution Panels - Basic, Switched, Metered. 1-3 phases. Up to 600V. NMC:N4F9qvHz11BHcc4nh1LCJFsrZhA1EWgVwj
|
|
|
zvisha
|
|
February 06, 2015, 03:23:37 PM |
|
Yes, I am sorry. I also have pending pull request from Luke which is probably good, but I do not have the time to make the integration (with all the release cycle, and testing on all the miner flavors). Especially in the web, which is not my strongest side, I am reluctant to change things But developers are welcome to do custom builds, and develop their own flavors if they feel it is interesting for them. Zvi, there's no reason to say "I'm sorry" In my opinion the code repositories should be used interactively. Some of the guys here are addicted enough to even code stuff for you (web or not), but unfortunately they can't integrate their stuff in the mainstream. That's why 2112 suggested a step back from day to day issues and look back in with a bigger perspective. You have a unique position being on the team since the first SP-T chip was being created. Since SP35 is almost out of stock, with the new chips (if any) one can try a new approach. It will be a pity to use the same (or slightly modified) chip optimization model since the SP10 units From a personal perspective I know it's hard to break routines and take new challenges when there is a multitude of small issues to be solved. However, those issues will always be there. Sometimes I'd like to code stuff instead of driving my team to do the coding Particularly when the team is small and busy. But they need guidance and planning while they do what they do best: code. They can't be efficient when they focus on both current issues and future planning. Someone needs to do the planning for them It's the "management" job security paradigm With the continuous increase in the number of SP-T models, just testing new firmware versions for all situations out there will eventually stress your team more than it can cope with. If I were you, I'd drop support for older models (I'm still happy with my SP10 running FW 1.5.8 and I'm not looking to upgrade it any longer). That will let you focus on the new ones. Also, try an "agile" way of running things. Let your guys focus on the improvements with the largest impact and with minimum work-hours .. the R&D supporting the firmware using the future line of chips should be one of those. By the way, no one needs a weekly FW update. It can be bi-weekly or even monthly. There's no time to upgrade and monitor the units (particularly in large numbers) for improvement, that quick. What it really boils out of it, is to sort out priorities, communicating the sort result to the team. And one can't do that properly when (s)he's always busy with small stuff. I believe that the fw logic is very close to optimum. Despite some things said here i have mathematical background, and i took part in both asic interface definition and implementation. i use greedy algorithm for finding the optimal working point not because it is simplest but because i believe it is best for this task. And i did some simulations and tests before reaching that conclusion. I see no point to rewrite any of the existing fws since i dont believe it will give higher performance by more then a couple of %. I would be happy to be proven wrong by someone from this forum )
|
|
|
|
GenTarkin
Legendary
Offline
Activity: 2450
Merit: 1002
|
|
February 06, 2015, 03:39:30 PM |
|
Yes, I am sorry. I also have pending pull request from Luke which is probably good, but I do not have the time to make the integration (with all the release cycle, and testing on all the miner flavors). Especially in the web, which is not my strongest side, I am reluctant to change things But developers are welcome to do custom builds, and develop their own flavors if they feel it is interesting for them. Zvi, there's no reason to say "I'm sorry" In my opinion the code repositories should be used interactively. Some of the guys here are addicted enough to even code stuff for you (web or not), but unfortunately they can't integrate their stuff in the mainstream. That's why 2112 suggested a step back from day to day issues and look back in with a bigger perspective. You have a unique position being on the team since the first SP-T chip was being created. Since SP35 is almost out of stock, with the new chips (if any) one can try a new approach. It will be a pity to use the same (or slightly modified) chip optimization model since the SP10 units From a personal perspective I know it's hard to break routines and take new challenges when there is a multitude of small issues to be solved. However, those issues will always be there. Sometimes I'd like to code stuff instead of driving my team to do the coding Particularly when the team is small and busy. But they need guidance and planning while they do what they do best: code. They can't be efficient when they focus on both current issues and future planning. Someone needs to do the planning for them It's the "management" job security paradigm With the continuous increase in the number of SP-T models, just testing new firmware versions for all situations out there will eventually stress your team more than it can cope with. If I were you, I'd drop support for older models (I'm still happy with my SP10 running FW 1.5.8 and I'm not looking to upgrade it any longer). That will let you focus on the new ones. Also, try an "agile" way of running things. Let your guys focus on the improvements with the largest impact and with minimum work-hours .. the R&D supporting the firmware using the future line of chips should be one of those. By the way, no one needs a weekly FW update. It can be bi-weekly or even monthly. There's no time to upgrade and monitor the units (particularly in large numbers) for improvement, that quick. What it really boils out of it, is to sort out priorities, communicating the sort result to the team. And one can't do that properly when (s)he's always busy with small stuff. I believe that the fw logic is very close to optimum. Despite some things said here i have mathematical background, and i took part in both asic interface definition and implementation. i use greedy algorithm for finding the optimal working point not because it is simplest but because i believe it is best for this task. And i did some simulations and tests before reaching that conclusion. I see no point to rewrite any of the existing fws since i dont believe it will give higher performance by more then a couple of %. I would be happy to be proven wrong by someone from this forum ) How about letting us have direct control over clocks & voltages?!
|
|
|
|
klondike_bar
Legendary
Offline
Activity: 2128
Merit: 1005
ASIC Wannabe
|
|
February 06, 2015, 03:45:55 PM |
|
Current firmware version: 2.6.12 shutdown button should have an yes and cancel question, not directly shutdown!
what can i do about this:
LOOP[0] OFF TO:0 (test serial failed or something) 0: disabled (serial problem) 1: disabled (serial problem)
I have a unit like that - i switched the two serial cables around and the issue moved to loop [2]. The issue therefore is my serial cable and not the hashing board. you might want to see if spondoolies will allow you to open the unit (may void warranty) to try reseating or swapping the ribbon cables.
|
|
|
|
mavericklm
|
|
February 06, 2015, 04:10:08 PM Last edit: February 06, 2015, 04:23:57 PM by mavericklm |
|
done already and now is full speed! it did this disabled (serial problem) sometimes and with a reboot and/or restart it was working again, but today it didn't want to start anymore for few hours. did the ribbon cable swap and now is ok. it might be only a bad contact between ribbon cable and the boards, maybe the other issue: i underclock the unit to 220w per loop and now i got no more lazy asic thx for your time
|
|
|
|
shacky
|
|
February 06, 2015, 04:33:02 PM |
|
Hi,
Yesterday I ordered a mini farm (#4521), and didn't recibe the tracking code, when you will ship it?
thanks
|
|
|
|
Tupsu
Legendary
Offline
Activity: 1218
Merit: 1003
|
|
February 06, 2015, 04:53:29 PM |
|
Hi,
Yesterday I ordered a mini farm (#4521), and didn't recibe the tracking code, when you will ship it?
thanks
Probably Sunday.
|
|
|
|
|