Bitcoin Forum
May 03, 2024, 04:39:47 PM *
News: Latest Bitcoin Core release: 27.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: [1]
  Print  
Author Topic: How are rejected shares related to mining intensity?  (Read 2433 times)
lorbas (OP)
Member
**
Offline Offline

Activity: 82
Merit: 10


View Profile
March 03, 2014, 10:49:27 PM
 #1

Hey there!

Could anybody please explain to my why I get fewer rejected shares when lowering intensity? I'm running a 6 card setup with 7950's and 1080MHz/1250MHZ clocks, cgminer 3.7.2 @ I20. When pool-mining - I figuered out that rejected shares are only produced when a new block is found. Then I mostly end up with a rejection rate of 06%-10%. I've tested like 10 pools now. Latency does not seem to be an issue.

Now people keep saying that you got to lower intensity when the rejection rate is too high. And I guess they might be right. Cause lowering my intensity actually results in fewer rejected shares. But apart from that - what's the reason for it? Shouldn't it be the other way round? Faster results = faster feedback to the pool = fewer rejections?!

I worked quite hard tweaking my setup, finding the perfect clock speeds, undervolts, bioses and bins for a rock solid, stable system - Now do I REALLY have to lower intensity/hashrate so that I don't end up with 10% wasted computes?
1714754387
Hero Member
*
Offline Offline

Posts: 1714754387

View Profile Personal Message (Offline)

Ignore
1714754387
Reply with quote  #2

1714754387
Report to moderator
1714754387
Hero Member
*
Offline Offline

Posts: 1714754387

View Profile Personal Message (Offline)

Ignore
1714754387
Reply with quote  #2

1714754387
Report to moderator
1714754387
Hero Member
*
Offline Offline

Posts: 1714754387

View Profile Personal Message (Offline)

Ignore
1714754387
Reply with quote  #2

1714754387
Report to moderator
BitcoinCleanup.com: Learn why Bitcoin isn't bad for the environment
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
1714754387
Hero Member
*
Offline Offline

Posts: 1714754387

View Profile Personal Message (Offline)

Ignore
1714754387
Reply with quote  #2

1714754387
Report to moderator
1714754387
Hero Member
*
Offline Offline

Posts: 1714754387

View Profile Personal Message (Offline)

Ignore
1714754387
Reply with quote  #2

1714754387
Report to moderator
1714754387
Hero Member
*
Offline Offline

Posts: 1714754387

View Profile Personal Message (Offline)

Ignore
1714754387
Reply with quote  #2

1714754387
Report to moderator
DeathAndTaxes
Donator
Legendary
*
Offline Offline

Activity: 1218
Merit: 1079


Gerald Davis


View Profile
March 03, 2014, 10:57:14 PM
Last edit: March 04, 2014, 07:14:05 PM by DeathAndTaxes
 #2

GPU operate in a "batch" mode.  They start hashing and attempt so many hashes before stopping to return results.  The higher the intensity the more hashes the card attempts in one batch but also the longer the batch takes.  The card only returns results at the end of the batch.  The higher the intensity (and longer the batch) the more shares which are computed but never "delivered" before the block changes and they become stale.  On the other hand the lower the intensity the more of the GPU time spent setting up and taking down batches which means a lower raw hashrate

You get paid for shares (or work) accepted so that is what you should be looking to maximize.  It is possible the accepted rate is still higher even with a higher rejected rate because the gross (raw) hashrate is higher.

Simplified example
100 shares gross w/ 1% rejected = 99 shares accepted
120 shares gross w/ 10% rejected = 102 shares accepted.

The later is more profitable.  Since you get "paid" for shares accepted that is the number you want to maximize.
Dalkore
Legendary
*
Offline Offline

Activity: 1330
Merit: 1026


Mining since 2010 & Hosting since 2012


View Profile WWW
March 03, 2014, 11:00:08 PM
 #3

GPU operate in a "batch" mode.  They start hashing and attempt so many hashes before stopping to return results.  The higher the intensity the more hashes the card attempts in one batch but also the longer the batch takes.  The card only returns results at the end of the batch.  The higher the intensity (and longer the batch) the more shares which are computed but are now "stale" when a block change occurs before the batch ends.

What matters is shares (or work) accepted.   It is possible the accepted rate is still higher even with a higher rejected rate because the gross rate is higher.

Simple example

100 shares gross w/ 1% rejected = 99 shares accepted
120 shares gross w/ 10% rejected = 102 shares accepted.

The later is more profitable.  All that matters is how many shares you get paid on.
]

Thank you D&T.  Love your responses.

Hosting: Low as $60.00 per KW - Link
Transaction List: jayson3 +5 - ColdHardMetal +3 - Nolo +2 - CoinHoarder +1 - Elxiliath +1 - tymm0 +1 - Johnniewalker +1 - Oscer +1 - Davidj411 +1 - BitCoiner2012 +1 - dstruct2k +1 - Philj +1 - camolist +1 - exahash +1 - Littleshop +1 - Severian +1 - DebitMe +1 - lepenguin +1 - StringTheory +1 - amagimetals +1 - jcoin200 +1 - serp +1 - klintay +1 - -droid- +1 - FlutterPie +1
lorbas (OP)
Member
**
Offline Offline

Activity: 82
Merit: 10


View Profile
March 04, 2014, 07:27:24 AM
 #4

It is possible the accepted rate is still higher even with a higher rejected rate because the gross rate is higher.

Thank you very much for the profound reply! I think I get it. What I don't get is, what the gross rate is all about. It't the first time I hear about it.
So your're saying I might be making more profit with more rejects/higher hashrate than by lowering intensity and reducing rejects.

Do all readers agree?  Wink
zureman90
Newbie
*
Offline Offline

Activity: 14
Merit: 0


View Profile
March 04, 2014, 08:07:14 AM
 #5

It is possible the accepted rate is still higher even with a higher rejected rate because the gross rate is higher.

Thank you very much for the profound reply! I think I get it. What I don't get is, what the gross rate is all about. It't the first time I hear about it.
So your're saying I might be making more profit with more rejects/higher hashrate than by lowering intensity and reducing rejects.

Do all readers agree?  Wink

What matters is the amount of accepted shares. So although you might have more rejected shares with a higher intensity, the thing that matters - the accepted shares - is also higher.
Do a quick calculation to see your effective hashrate (in previous posts called 'gross rate' and choose what's best:

effective hashrate = raw hashrate x percentage of accepted shares

with:
raw hashrate = what miner reports
percentage of accepted shares = 100% - percentage of rejected shares
lorbas (OP)
Member
**
Offline Offline

Activity: 82
Merit: 10


View Profile
March 05, 2014, 10:59:51 PM
Last edit: March 05, 2014, 11:12:16 PM by lorbas
 #6


You get paid for shares (or work) accepted so that is what you should be looking to maximize.  It is possible the accepted rate is still higher even with a higher rejected rate because the gross (raw) hashrate is higher. [...] Since you get "paid" for shares accepted that is the number you want to maximize.

So if that's correct - I conclude that for people like me, who are mining with high intensity and who are therefore taking a higher risk of getting rejects, pool-mining a coin with a higher difficulty (more time until a block is found) are better of, cause it results in fewer rejects.

Correct?

Since you get "paid" for shares accepted that is the number you want to maximize.

When setting a worker's  diff in a pool to a low valule, let's say 32, the number of accepted shares is way higer then with a worker diff of 1024. So I assume you get not only paid for the number of accepted shares but also the size of the batch. Otherwise all people would be mining at diff 32, right?
DeathAndTaxes
Donator
Legendary
*
Offline Offline

Activity: 1218
Merit: 1079


Gerald Davis


View Profile
March 05, 2014, 11:23:13 PM
 #7

So if that's correct - I conclude that for people like me, who are mining with high intensity and who are therefore taking a higher risk of getting rejects, pool-mining a coin with a higher difficulty (more time until a block is found) are better of, cause it results in fewer rejects.

Correct?

No it is irrelevant.  Difficulty doesn't affect the time between block changes.


Quote
When setting a worker's  diff in a pool to a low valule, let's say 32, the number of accepted shares is way higer then with a worker diff of 1024. So I assume you get not only paid for the number of accepted shares but also the size of the batch. Otherwise all people would be mining at diff 32, right?

Share difficulty, min pool share difficulty, and block difficulty all have nothing to do with intensity (batch size).

Intensity defines how long the card will run (in hashes) before it ends the batch, drops off results, and picks up new instructions.

You are right (but using wrong words) in that pools pay based on the total difficulty of the shares accepted not the raw number of shares.  A miner with lower difficulty will produce more shares but each share is worth less and a miner running at higher difficulty will produce less shares but each share is worth more.   As an analogy someone depositing one $100 bill into the register gets no more credit than someone else depositing one hundred $1 bills in the register. This has nothing to do with intensity though. 

Pages: [1]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!