DiabloMiner GPU Miner - page 58. | Bitcointalksearch.org

whackedspinach

newbie

Activity: 11

Merit: 0

Quote from: DiabloD3 on June 03, 2011, 07:51:51 PM

Quote from: whackedspinach on June 03, 2011, 06:58:20 PM

Code:

[6/3/11 6:49:05 PM] Block 10 found on Barts (#2)
[6/3/11 6:49:09 PM] Block 11 found on Barts (#1)
[6/3/11 6:49:29 PM] Block 12 found on Barts (#2)
[6/3/11 6:49:40 PM] Block 13 found on Barts (#1)
[6/3/11 6:49:44 PM] Block 14 found on Barts (#2)
[6/3/11 6:50:25 PM] Block 15 found on Barts (#2)
[6/3/11 6:50:43 PM] Block 16 found on Barts (#2)
[6/3/11 6:50:57 PM] Block 17 found on Barts (#1)

So, if DM is outputting stuff like this, then are the blocks defiantly being found by the cards that are in parentheses? What I mean is that Barts (#2) doesn't seem to be accessible by certain aticonfig options, and my hash rate seems to be that of a single GPU. Is it possible that all blocks are being completed by Barts #1 but being labeled otherwise?

Nope. Sure you have the right speed in mind? A 6870 should be doing somewhere around 270 or so at stock speeds.

I have two 6850s and I'm getting 230 overclocked (to a core clock of 900). I can't use aticonfig to set any setting for Barts #2, yet DM says it is finding blocks, so I'm confused if it is contributing to the hash meter or not.

DiabloD3

legendary

Activity: 1162

Merit: 1000

DiabloMiner author

Quote from: melanarchy on June 03, 2011, 06:58:48 PM

Quote from: melanarchy on June 03, 2011, 01:21:49 PM

Okay I'll try Diablo with a smaller pool that has been more reliable for me in the past and let you know if that helps.

Quote from: DiabloD3 on June 03, 2011, 01:12:22 PM

Like I've said to others, I believe there is either a widespread bug in how some pools operate (such as giving the wrong diff target out, or incorrectly checking hashes (such as H == 0 only instead of also G <= target)), or Pheonix is incorrectly reporting that shares have been accepted when they haven't.

For me, on Deepbit, I get two or three stale shares for every 1000 give or take.

I connected to a small pool I often use with Diablo. It kept up reporting 304Mhash/s but after several hours the pool reported that of the 1231 requested blocks Diablo had only been able to return 854, for an efficiency of 69%.

With phoenix my efficiency is upwards of 90% even though my Mhash rate is slower by 5. I don't know enough about the long polling process or the way the rpcs work to identify if this is a bug with my pool or the miner. But my effective hash rate is certainly better with phoenix.

You're suffering from the bug. It eats about 1/3rd of hash attempts.

Although, interestingly, not using vectors doesn't trigger it.

Sigh. I hate bugs.

DiabloD3

legendary

Activity: 1162

Merit: 1000

DiabloMiner author

Quote from: brunoshady on June 03, 2011, 07:05:17 PM

and why diablo miner shows xxxx/yyyyy khash ?

what does the yyyy means?

15 second average/forever average

DiabloD3

legendary

Activity: 1162

Merit: 1000

DiabloMiner author

Quote from: brunoshady on June 03, 2011, 07:03:35 PM

how can I check my sdk version?

where can I see it?

It spits it out on startup now.

DiabloD3

legendary

Activity: 1162

Merit: 1000

DiabloMiner author

Quote from: whackedspinach on June 03, 2011, 06:58:20 PM

Code:

[6/3/11 6:49:05 PM] Block 10 found on Barts (#2)
[6/3/11 6:49:09 PM] Block 11 found on Barts (#1)
[6/3/11 6:49:29 PM] Block 12 found on Barts (#2)
[6/3/11 6:49:40 PM] Block 13 found on Barts (#1)
[6/3/11 6:49:44 PM] Block 14 found on Barts (#2)
[6/3/11 6:50:25 PM] Block 15 found on Barts (#2)
[6/3/11 6:50:43 PM] Block 16 found on Barts (#2)
[6/3/11 6:50:57 PM] Block 17 found on Barts (#1)

So, if DM is outputting stuff like this, then are the blocks defiantly being found by the cards that are in parentheses? What I mean is that Barts (#2) doesn't seem to be accessible by certain aticonfig options, and my hash rate seems to be that of a single GPU. Is it possible that all blocks are being completed by Barts #1 but being labeled otherwise?

Nope. Sure you have the right speed in mind? A 6870 should be doing somewhere around 270 or so at stock speeds.

brunoshady

sr. member

Activity: 262

Merit: 250

Dubs Get

and why diablo miner shows xxxx/yyyyy khash ?

what does the yyyy means?

brunoshady

sr. member

Activity: 262

Merit: 250

Dubs Get

how can I check my sdk version?

where can I see it?

melanarchy

newbie

Activity: 22

Merit: 0

Quote from: melanarchy on June 03, 2011, 01:21:49 PM

Okay I'll try Diablo with a smaller pool that has been more reliable for me in the past and let you know if that helps.

Quote from: DiabloD3 on June 03, 2011, 01:12:22 PM

Like I've said to others, I believe there is either a widespread bug in how some pools operate (such as giving the wrong diff target out, or incorrectly checking hashes (such as H == 0 only instead of also G <= target)), or Pheonix is incorrectly reporting that shares have been accepted when they haven't.

For me, on Deepbit, I get two or three stale shares for every 1000 give or take.

I connected to a small pool I often use with Diablo. It kept up reporting 304Mhash/s but after several hours the pool reported that of the 1231 requested blocks Diablo had only been able to return 854, for an efficiency of 69%.

With phoenix my efficiency is upwards of 90% even though my Mhash rate is slower by 5. I don't know enough about the long polling process or the way the rpcs work to identify if this is a bug with my pool or the miner. But my effective hash rate is certainly better with phoenix.

whackedspinach

newbie

Activity: 11

Merit: 0

Code:

[6/3/11 6:49:05 PM] Block 10 found on Barts (#2)
[6/3/11 6:49:09 PM] Block 11 found on Barts (#1)
[6/3/11 6:49:29 PM] Block 12 found on Barts (#2)
[6/3/11 6:49:40 PM] Block 13 found on Barts (#1)
[6/3/11 6:49:44 PM] Block 14 found on Barts (#2)
[6/3/11 6:50:25 PM] Block 15 found on Barts (#2)
[6/3/11 6:50:43 PM] Block 16 found on Barts (#2)
[6/3/11 6:50:57 PM] Block 17 found on Barts (#1)

So, if DM is outputting stuff like this, then are the blocks defiantly being found by the cards that are in parentheses? What I mean is that Barts (#2) doesn't seem to be accessible by certain aticonfig options, and my hash rate seems to be that of a single GPU. Is it possible that all blocks are being completed by Barts #1 but being labeled otherwise?

[Tycho]

hero member

Activity: 742

Merit: 500

Quote from: DiabloD3 on June 03, 2011, 11:52:36 AM

Let me repeat: the Deepbit hashmeter is useless. It does not reflect anything useful. I've had it say I'm doing 500 mhash consistently for about an hour, and I do around 300.

I call it "the luck meter" for a reason.
It may not reflect your real hashing speed correctly, but your reward will be the same as if you were hashing with that displayed speed and "normal" luck.
So if you saw 500 MH/s on the luck meter, then you were 200 MH/s more lucky and got 66% more BTCs.

organofcorti

donator

Activity: 2058

Merit: 1007

Poor impulse control.

Quote

Edit: double fuck, its doing it here too.

No bug for me using DM dl'd on 31-5. I'm on LinuxCoin, AMD APP SDK v2.4, ATI catalyst & drivers 11.3 and one solitary 6990. I get arouns 700000Mhps which is similar to what Eligius is giving me:

Code:

This user has submitted 1,716 shares in the last 3 hours. This represents a contribution in average of 682.42 Mhashes/sec to the pool.

BTW, (in case this bug doesn't have you busy as a blue arsed fly) I haven't been able to figure out how to run DM from cron, either by calling a script or DM directly. It fails with a bunch of crap about not being able to do something inexplicable in java. I guess I'm not adding the correct folder to PATH in cron? Or something? Any ideas? I'll post the error message later if it helps.

DiabloD3

legendary

Activity: 1162

Merit: 1000

DiabloMiner author

Using the log vonshmitt sent me, the people suffering from this bug are losing about 1/3rd shares (rejected and accepted combined).

I wonder if I missed additional bugs in mtrlt's patch.

Edit: double fuck, its doing it here too.

DiabloD3

legendary

Activity: 1162

Merit: 1000

DiabloMiner author

Quote from: nyargh on June 03, 2011, 02:18:08 PM

Quote from: DiabloD3 on June 03, 2011, 11:52:36 AM

I cannot reproduce this bug.

That's fine, there are other miners.

Well, if the bug happens to be on the pool side (ie, pools are not properly checking hashes), then not only is that exploitable, it would be blocked by the pool software I'm writing (which checks share validity fully).

If the bug IS in my miner, I would love to find out what exactly triggers it so I can fix it, but I just don't think it is.

DiabloD3

legendary

Activity: 1162

Merit: 1000

DiabloMiner author

Quote from: GODLIKE on June 03, 2011, 02:12:26 PM

Quote from: GODLIKE on June 02, 2011, 05:10:09 PM

Quote from: DiabloD3 on June 02, 2011, 04:27:32 PM

Either you do not have OpenCL drivers installed, or your device does not support OpenCL.

Duh I didn't read anything about the need of OpenCL anywhere... it should be included with ATI drivers right? So it should be installed already.

Nonetheless, it's not workin, what should I do?

If you're on Windows, and have installed a recent version of Catalyst (11.4 and up), it installs the OpenCL runtime unless you tell it not to.

What hardware do you have?

nyargh

newbie

Activity: 25

Merit: 0

Quote from: DiabloD3 on June 03, 2011, 11:52:36 AM

I cannot reproduce this bug.

That's fine, there are other miners.

GODLIKE

hero member

Activity: 546

Merit: 500

LOL what you looking at?

Quote from: GODLIKE on June 02, 2011, 05:10:09 PM

Quote from: DiabloD3 on June 02, 2011, 04:27:32 PM

Either you do not have OpenCL drivers installed, or your device does not support OpenCL.

Duh I didn't read anything about the need of OpenCL anywhere... it should be included with ATI drivers right? So it should be installed already.

Nonetheless, it's not workin, what should I do?

melanarchy

newbie

Activity: 22

Merit: 0

Okay I'll try Diablo with a smaller pool that has been more reliable for me in the past and let you know if that helps.

Quote from: DiabloD3 on June 03, 2011, 01:12:22 PM

Like I've said to others, I believe there is either a widespread bug in how some pools operate (such as giving the wrong diff target out, or incorrectly checking hashes (such as H == 0 only instead of also G <= target)), or Pheonix is incorrectly reporting that shares have been accepted when they haven't.

For me, on Deepbit, I get two or three stale shares for every 1000 give or take.

DiabloD3

legendary

Activity: 1162

Merit: 1000

DiabloMiner author

Quote from: melanarchy on June 03, 2011, 01:03:08 PM

So I've been mining with phoenix and phatk and I get 299Mhash/sec pretty consistently (VECTORS BFI_INT AGGRESSION=12 WORKSIZE=128). I tried Diablo and it lets me push it up to around 304Mhash/sec (-v 19 -w 192) (Headless linux miner, 6870 1000MHz core 920Mhz memory). That does seem higher, but with phoneix I get an average of 98% accepted blocks where as with Diablo I'm seeing 90% accepted or below. That's a big difference. I'd love the 5 Mhash/sec more, but the drop in efficiency makes my actual throughput lower.

Any suggestions?

Like I've said to others, I believe there is either a widespread bug in how some pools operate (such as giving the wrong diff target out, or incorrectly checking hashes (such as H == 0 only instead of also G <= target)), or Pheonix is incorrectly reporting that shares have been accepted when they haven't.

For me, on Deepbit, I get two or three stale shares for every 1000 give or take.

melanarchy

newbie

Activity: 22

Merit: 0

So I've been mining with phoenix and phatk and I get 299Mhash/sec pretty consistently (VECTORS BFI_INT AGGRESSION=12 WORKSIZE=128). I tried Diablo and it lets me push it up to around 304Mhash/sec (-v 19 -w 192) (Headless linux miner, 6870 1000MHz core 920Mhz memory). That does seem higher, but with phoneix I get an average of 98% accepted blocks where as with Diablo I'm seeing 90% accepted or below. That's a big difference. I'd love the 5 Mhash/sec more, but the drop in efficiency makes my actual throughput lower.

Any suggestions?

DiabloD3

legendary

Activity: 1162

Merit: 1000

DiabloMiner author

Quote from: gentakin on June 03, 2011, 08:50:49 AM

Quote from: DiabloD3 on June 02, 2011, 05:33:59 PM

If its the bug I'm thinking it is, then you will get 50 attempts, but not 50 shares. If its the bfi_int harmless driver bug, then you will get 50 shares, but way more than 50 attempts.

So I started up diablo and after 14 minutes:
* 42 attempts
* 31 shares
* 273378/271347 khash/sec | ghash: 226,4

Not sure if this will help you track down what is going on with my setup. It looks like it's the "bad" bug, not the harmless bug. That would explain why the deepbit hashmeter never ever reached more than about 200mhash/s.
Maybe I'm just trying to get more out of my hardware than I should, thus really causing a "possible driver or hardware issue"?

edit: I also modified diablominer so it will send attempts even though H != 0, but that didn't work of course (ERROR: Connection failed: Bitcoin returned error message: Wrong data: checkWork: checkHash wrong).

Let me repeat: the Deepbit hashmeter is useless. It does not reflect anything useful. I've had it say I'm doing 500 mhash consistently for about an hour, and I do around 300.

And yes, don't try to send failed hashes. Not a good idea.

42 attempts producing 31 shares honestly sounds like the hardware is overheating or your machine is unstable. As long as your GPU temp is below 85c, this shouldn't happen unless the card sucks (say, Powercolor, Diamond, non-reference XFX, etc) or your PSU can't handle the load.

I cannot reproduce this bug.

Topic: DiabloMiner GPU Miner - page 58. (Read 866596 times)