And I believe you probably meant "empirical evidence". 
Lol. May be a sign of obsessing on grading papers.
Topic: [4+ EH] Slush Pool (slushpool.com); Overt AsicBoost; World First Mining Pool - page 332. (Read 4382810 times)
May 01, 2014, 04:38:11 PM
Lol. May be a sign of obsessing on grading papers.
May 01, 2014, 01:53:34 PM
Come on slushies lets find a block!!
May 01, 2014, 10:52:20 AM
Bear in mind that Slush does not have direct access to your miner's hash rate, but has to derive it from back-calculation based on the work performed. Your quoted variance of 15% is not something I would personally worry about.
I know they're not going to be the same, but I hadn't usually seen so much as 15% variance. Again, I'm not too worried about it, and I'll know more only after the block is solved.
And I believe you probably meant "empirical evidence".
Quite right. Thank you very much. I'm always happy to be reminded when I get grammar/wording/spelling wrong. I need to continue to improve if I want to have the ability to correct others. ;-)
May 01, 2014, 10:19:47 AM
Is anyone else seeing some strange data in the estimated reward and their current scores? I have a current score of around 6000 with over 8700 shares and my estimated reward is about 2% of what I normally see for the same number of shares. My hashrate appears normal.
Any ideas or explanations?
Any ideas or explanations?
Just a glitch. Happens from time to time. If you think its your vardiff then run a proxy.
May 01, 2014, 10:00:00 AM
You're right. I see that. However, the current round has been running for 16.5 hours. I brought one of my workers back online about 11 hours ago. And, the Mhash/s reported for that worker is somewhere around 85% of what my worker is reporting. And, in the past, this same worker has shown a hash rate on the pool of the same as it's reporting.
I won't be able to tell for sure until the block finishes and the next round starts, but imperial evidence suggests that the reported is _not_ a 60-minute average. I'll keep an eye on it, but.
Bear in mind that Slush does not have direct access to your miner's hash rate, but has to derive it from back-calculation based on the work performed. Your quoted variance of 15% is not something I would personally worry about.I won't be able to tell for sure until the block finishes and the next round starts, but imperial evidence suggests that the reported is _not_ a 60-minute average. I'll keep an eye on it, but.
And I believe you probably meant "empirical evidence".
May 01, 2014, 09:32:20 AM
It's not true. Column name is "Mhash/s*". And below that is "* Average hash rate in last 60 minutes. Calculated each 5 minutes.". I really think that this means, that it's not calculated over whole current round 
You're right. I see that. However, the current round has been running for 16.5 hours. I brought one of my workers back online about 11 hours ago. And, the Mhash/s reported for that worker is somewhere around 85% of what my worker is reporting. And, in the past, this same worker has shown a hash rate on the pool of the same as it's reporting.
I won't be able to tell for sure until the block finishes and the next round starts, but imperial evidence suggests that the reported is _not_ a 60-minute average. I'll keep an eye on it, but.
May 01, 2014, 02:38:46 AM
I have a question that is probably already answered somewhere, but, I haven't been able to find the answer.
The "MH/s" shown on the "My Account" page, for each worker. Over what time period is that? I assume it's an average over some time period, but, what is the time period?
Thanks....
The "MH/s" shown on the "My Account" page, for each worker. Over what time period is that? I assume it's an average over some time period, but, what is the time period?
Thanks....
current round
It's not true. Column name is "Mhash/s*". And below that is "* Average hash rate in last 60 minutes. Calculated each 5 minutes.". I really think that this means, that it's not calculated over whole current round
April 30, 2014, 06:52:06 PM
Is anyone else seeing some strange data in the estimated reward and their current scores? I have a current score of around 6000 with over 8700 shares and my estimated reward is about 2% of what I normally see for the same number of shares. My hashrate appears normal.
Any ideas or explanations?
With a very modest hashrate (about 7.5 GH/s split evenly between Slush and BTCguild) I see large variations in Slush. This seems to be due in no small part to over-aggressiveness on the part of Vardiff: it appears to go in cycles over about 20-30 minutes, dropping from difficulty 4 to 1, then up to 2, then up to 3 or 4, then back down to 1, resulting in big swings in the estimated reward. BTCguild, meanwhile, remains steady on the minimum of 2. Sod's Law decrees that Slush will often solve a block just when the estimated reward is at its low point, resulting in variance of over 50% even on reasonably long rounds when one would expect some consistency.Any ideas or explanations?
It seems to me that Vardiff is too sensitive to sort-term changes; while this may be a useful attribute in the first few minutes after establishing a connection, it becomes a liability once the worker has settled down.
Maybe vardiff needs to tone itself down after N shares or X minutes of runtime?
April 30, 2014, 05:55:30 PM
Is anyone else seeing some strange data in the estimated reward and their current scores? I have a current score of around 6000 with over 8700 shares and my estimated reward is about 2% of what I normally see for the same number of shares. My hashrate appears normal.
Any ideas or explanations?
With a very modest hashrate (about 7.5 GH/s split evenly between Slush and BTCguild) I see large variations in Slush. This seems to be due in no small part to over-aggressiveness on the part of Vardiff: it appears to go in cycles over about 20-30 minutes, dropping from difficulty 4 to 1, then up to 2, then up to 3 or 4, then back down to 1, resulting in big swings in the estimated reward. BTCguild, meanwhile, remains steady on the minimum of 2. Sod's Law decrees that Slush will often solve a block just when the estimated reward is at its low point, resulting in variance of over 50% even on reasonably long rounds when one would expect some consistency.Any ideas or explanations?
It seems to me that Vardiff is too sensitive to sort-term changes; while this may be a useful attribute in the first few minutes after establishing a connection, it becomes a liability once the worker has settled down.
April 30, 2014, 02:00:47 PM
Your profile page breaks down stats per worker.
Right, on the My Account page. That's what I was asking about originally. But, that only shows an average for the worker across the whole of the current round. I'd like to see a 20 minute, or even 1 hour, average. So that I can see any effect of tuning in my workers without having to wait more than a full round.
I believe that Slush implied on the Facebook page that functionality like this is coming "Soon (TM)".
April 30, 2014, 01:27:43 PM
Sorry everyone for starting such a heated discussion over repairs.
I just finished contacting my second local college and found some interesting things about the bitmain s1 antiminer.
The motherboards are NOT normal boards and can not be soldered on. Heat will destroy them. Guess this is the future of motherboards now. They are only one sided boards with everything added in layers over layers using some form of adheasive/electrical process to make things stick and connect.
What happened on mine (and just realized it was a brand new one just got in 10 days ago) was the item pulled away from the motherboard completely. Just fell off. Piss poor workmanship all done by computers.
Of course I purchased several off ebay as new and can't return them so I'm out, my bad I guess. Anyway bitcoin is always a high risk venture and always will be for the time being.
Guess will be selling it for parts for best I can get. Just letting everyone know the motherboards are not like TV or computer boards etc. That is why they don't repair them. If they work good enough to pass test that's all they care. I have 7 computers and workstations and only 2 have similar motherboards where items (at least larger ones) can't be replaced without microscopes and some sort of fancy adhearing processs etc.
If you bought it off fleabay, I doubt it was new, doesn't sound like it in the end... anyway, I broke some capacitors off mine and a guy at my local computer shop just soldered them back on for me, nothing fancy at all... I just finished contacting my second local college and found some interesting things about the bitmain s1 antiminer.
The motherboards are NOT normal boards and can not be soldered on. Heat will destroy them. Guess this is the future of motherboards now. They are only one sided boards with everything added in layers over layers using some form of adheasive/electrical process to make things stick and connect.
What happened on mine (and just realized it was a brand new one just got in 10 days ago) was the item pulled away from the motherboard completely. Just fell off. Piss poor workmanship all done by computers.
Of course I purchased several off ebay as new and can't return them so I'm out, my bad I guess. Anyway bitcoin is always a high risk venture and always will be for the time being.
Guess will be selling it for parts for best I can get. Just letting everyone know the motherboards are not like TV or computer boards etc. That is why they don't repair them. If they work good enough to pass test that's all they care. I have 7 computers and workstations and only 2 have similar motherboards where items (at least larger ones) can't be replaced without microscopes and some sort of fancy adhearing processs etc.
April 30, 2014, 11:47:07 AM
Is anyone else seeing some strange data in the estimated reward and their current scores? I have a current score of around 6000 with over 8700 shares and my estimated reward is about 2% of what I normally see for the same number of shares. My hashrate appears normal.
Any ideas or explanations?
Any ideas or explanations?
Happens sometimes, should correct itself pretty quick. It's only a problem if the round just end's when this happens and even then slush corrects this...
April 30, 2014, 11:11:03 AM
Is anyone else seeing some strange data in the estimated reward and their current scores? I have a current score of around 6000 with over 8700 shares and my estimated reward is about 2% of what I normally see for the same number of shares. My hashrate appears normal.
Any ideas or explanations?
Any ideas or explanations?
April 30, 2014, 10:13:48 AM
Your profile page breaks down stats per worker.
Right, on the My Account page. That's what I was asking about originally. But, that only shows an average for the worker across the whole of the current round. I'd like to see a 20 minute, or even 1 hour, average. So that I can see any effect of tuning in my workers without having to wait more than a full round.
April 29, 2014, 11:29:55 PM
current round
Thanks. I thought that might be the case. Makes it hard to see variation in any changes I make in the config of my miner, since rounds sometimes last most of a day, but understandable. Is there any break-down of what work is being contributed by workers elsewhere in the interface?
Your profile page breaks down stats per worker.
April 29, 2014, 04:01:25 PM
current round
Thanks. I thought that might be the case. Makes it hard to see variation in any changes I make in the config of my miner, since rounds sometimes last most of a day, but understandable. Is there any break-down of what work is being contributed by workers elsewhere in the interface?
April 29, 2014, 03:49:48 PM
I have a question that is probably already answered somewhere, but, I haven't been able to find the answer.
The "MH/s" shown on the "My Account" page, for each worker. Over what time period is that? I assume it's an average over some time period, but, what is the time period?
Thanks....
The "MH/s" shown on the "My Account" page, for each worker. Over what time period is that? I assume it's an average over some time period, but, what is the time period?
Thanks....
current round
April 29, 2014, 03:30:46 PM
I have a question that is probably already answered somewhere, but, I haven't been able to find the answer.
The "MH/s" shown on the "My Account" page, for each worker. Over what time period is that? I assume it's an average over some time period, but, what is the time period?
Thanks....
The "MH/s" shown on the "My Account" page, for each worker. Over what time period is that? I assume it's an average over some time period, but, what is the time period?
Thanks....
April 29, 2014, 03:25:30 PM
Some miners are sensitive to the network delays and submitting several shares per second is causing them to fail getting to higher diff - had the same problem some time ago and using proxy to keep the diff between (few) restarts helped me to 'climb' to higher diff where the miner is already happy to continue
April 29, 2014, 03:09:09 PM
I just got a Dragon 1 Ths miner and I'm struggling to get it to work here.
[...]
I have a feeling this is a minimum difficulty setting as Guild allows me to set 1024 which lets the miner work.
There is no such setting at Slush though.
[...]
[...]
I have a feeling this is a minimum difficulty setting as Guild allows me to set 1024 which lets the miner work.
There is no such setting at Slush though.
[...]
I have a CoinTerra running at Slush's pool, and it does suck for a little bit when it starts up at a difficulty of 3. But it doesn't take very long for it to vardiff itself up to 1100 or 1200. Is vardiff not working for you? Is that the problem you're seeing, or is it some other connection problem?
Jump to: