CCminer(SP-MOD) Modded GPU kernels.

[Nikolaj]

djm, I've sent you a pm about it.

Thanks

[Nikolaj]

the profit calculators between suprnova and pool.mn are identical, what I cannot understand it's the inhability to reach those value on a 20% lower average. The loss it's too significant to consider a simple hashrate variation.

A 1080 should gain, right now, in 24h 16 coins, for a total of 0.02BTC, but you can obtain in the end just a bit more than the half.

Weird behaviour.

ps: no, i'm not talking about accepted share variations, pool efficiency on suprnova it's more than 95%. If we consider the net hashrate and the difficulty it makes no sense.

[Amph]

The last 24hours. 1 GB of lbry payed out around $38.352 (http://yiimp.ccminer.org/ on the column to right in the library column you see the actual payout for 1 ghash)

So if your miner mined at 78MHASH you would earn 3 bucks.
 
Djm34's 2 gtx 980's is only hashing @ 39MHASH each.


To reach 10$ per card you need 260MHASH. And this is doable on the 1080 with proper coding.

given the real earnings a gtx 1080 on yiimp at 250MH/s should get 5.9$/Day, gross.

Suprnova calculator it's way off. With an hashrate of 940MH, yeasterday on suprnova it earned 35 coins..

Something's wrong there, as always.

that was my suspect as well supernova estimate seems wrong or not on par with the block features of adding 1 coin on the reward for each additional 100 blocks

gtx 1070 is doing 0.007-0.01 btc/day with public available open-sourced miner on public pool. What is the reason to mess with libry?

where is the open source miner?

damn told before: a gtx 980 does around 190MH/s

$7.3 per day on the old Horse. Not bad.

yeah but consumption is prolly very bad in that case, not below 200w for sure

don't remember but I think the 980 is only 180W (running at 120% tdp)... that makes it mostly below 200W

probably the 1070 is not fully optimized, if a 980 cna do 190MH, a 1070 should be above 200MH

[Nikolaj]

If a 980 can really reach 190, it's not with a general MINER, but a testing one from the main dev.

With the current binaries, 150 with an high OC it's doable. 220 with a 1070oc it's perfectly in line.

[bensam1231]

The last 24hours. 1 GB of lbry payed out around $38.352 (http://yiimp.ccminer.org/ on the column to right in the library column you see the actual payout for 1 ghash)

So if your miner mined at 78MHASH you would earn 3 bucks.
 
Djm34's 2 gtx 980's is only hashing @ 39MHASH each.


To reach 10$ per card you need 260MHASH. And this is doable on the 1080 with proper coding.

given the real earnings a gtx 1080 on yiimp at 250MH/s should get 5.9$/Day, gross.

Suprnova calculator it's way off. With an hashrate of 940MH, yeasterday on suprnova it earned 35 coins..

Something's wrong there, as always.

Weird, I'm earning about on par on Suprnova with yiimp. -3%

The 'calculator', if you're referring to how much it says you're going to earn in the pool tab is estimated based on current pool income and network difficulty. Some coins fluctuate a LOT. Each pool will give you a different estimate.

I was mining on MNpool and one of the days I caught it, I earned about 50% of what I should, even though the pool wasn't having bad luck. There is a earning tab you can check your daily earnings and hourly earnings on.

Earnings tab is ACTUAL where pool tab is ESTIMATE.

[grrrgrrr]

The last 24hours. 1 GB of lbry payed out around $38.352 (http://yiimp.ccminer.org/ on the column to right in the library column you see the actual payout for 1 ghash)

So if your miner mined at 78MHASH you would earn 3 bucks.
 
Djm34's 2 gtx 980's is only hashing @ 39MHASH each.


To reach 10$ per card you need 260MHASH. And this is doable on the 1080 with proper coding.

given the real earnings a gtx 1080 on yiimp at 250MH/s should get 5.9$/Day, gross.

Suprnova calculator it's way off. With an hashrate of 940MH, yeasterday on suprnova it earned 35 coins..

Something's wrong there, as always.

Weird, I'm earning about on par on Suprnova with yiimp. -3%

The 'calculator', if you're referring to how much it says you're going to earn in the pool tab is estimated based on current pool income and network difficulty. Some coins fluctuate a LOT. Each pool will give you a different estimate.

I was mining on MNpool and one of the days I caught it, I earned about 50% of what I should, even though the pool wasn't having bad luck. There is a earning tab you can check your daily earnings and hourly earnings on.

Earnings tab is ACTUAL where pool tab is ESTIMATE.

Pools usually use difficulty of the last 24 hours to estimate your earnings. The estimation today should in theory be consistent with how much you earn yesterday.

As to today's lbc earnings, with 800 Mh/s you should expect 28 usd.

That estimation is far from telling the whole story tho... I'm recently doing algorithm switching. There are hidden gems like UFO coin that sometimes earns you a $ in minutes but nothing for the rest of the day. It whacks coins with difficulty algorithms that suck are too sensitive.

[bensam1231]

The last 24hours. 1 GB of lbry payed out around $38.352 (http://yiimp.ccminer.org/ on the column to right in the library column you see the actual payout for 1 ghash)

So if your miner mined at 78MHASH you would earn 3 bucks.
 
Djm34's 2 gtx 980's is only hashing @ 39MHASH each.


To reach 10$ per card you need 260MHASH. And this is doable on the 1080 with proper coding.

given the real earnings a gtx 1080 on yiimp at 250MH/s should get 5.9$/Day, gross.

Suprnova calculator it's way off. With an hashrate of 940MH, yeasterday on suprnova it earned 35 coins..

Something's wrong there, as always.

Weird, I'm earning about on par on Suprnova with yiimp. -3%

The 'calculator', if you're referring to how much it says you're going to earn in the pool tab is estimated based on current pool income and network difficulty. Some coins fluctuate a LOT. Each pool will give you a different estimate.

I was mining on MNpool and one of the days I caught it, I earned about 50% of what I should, even though the pool wasn't having bad luck. There is a earning tab you can check your daily earnings and hourly earnings on.

Earnings tab is ACTUAL where pool tab is ESTIMATE.

Pools usually use difficulty of the last 24 hours to estimate your earnings. The estimation today should in theory be consistent with how much you earn yesterday.

As to today's lbc earnings, with 800 Mh/s you should expect 28 usd.

That estimation is far from telling the whole story tho... I'm recently doing algorithm switching, that is switching to the most profitable coin based on difficulty. There are hidden gems like UFO coin that sometimes earns you a couple $$ in minutes and nothing for the rest of the day. DGB and AUR can be very rewarding 50% of the time. Algorithm switching is profitable for coins with difficulty algorithms that suck.

Aye, and your hashrate fluctuates, so unless you know your absolute average on the pool, it's going to change a lot on top of hourly earnings fluctuating for the last 24 hours (bad hours and good hours).

Algo switching doesn't work nearly as well as you think it will due to it not being able to calculate your influence on a coins difficulty. That would have to be something really advanced and not something like minercontrol. Which is why you see some people dumping their hashrate and immediately crashing a coin (and of course getting almost no rewards for doing so). Depending on how fast the coin adjusts to hashrate. Some coins have really good difficulty adjustments, while others don't. Also something stupid like minercontrol doesn't take into account. Difficulty increases don't need to be linear.

I'm sure there is a big miner operation that actually has someone with expertise in coin development who takes all of this into account. Ideal solution really has nothing to do with hoping from coin to coin, but rather mining all relevant coins with just enough hashrate to profit from a smaller coin. Spreading the hash instead of bombing coins, but once again something that takes expertise and time to program.


Wolf0s miner finally caught up with Epsolyn's. Hope Epsolyn is still improving his.

Been working on LBRY some more...

Results (NSFW): https://ottrbutt.com/miner/wolflbry-07172016.png

160 on a 290x
240 on a Fury

[Amph]

again it's easy to do 240MH on fury when they consume 270w, not impressed

currently you can do 220 with a 1070 at half the wattage, if someone can check the utilizzation on lbry, if it is not 99% then there are probably some optimizations to do, otherwise not so much

i would like to see wolfo working on nivida some times, to see what he can really pull off, or if this is because you can do more with amd than nvidia...something is telling me that it is the latter...

[bensam1231]

again it's easy to do 240MH on fury when they consume 270w, not impressed

currently you can do 220 with a 1070 at half the wattage, if someone can check the utilizzation on lbry, if it is not 99% then there are probably some optimizations to do, otherwise not so much

i would like to see wolfo working on nivida some times, to see what he can really pull off, or if this is because you can do more with amd than nvidia...something is telling me that it is the latter...

AMD doesn't have the luxuries that Nvidia has for developers, which is one reason I liked it in the beginning. But what AMD does offer is unprecedented control, which I exploited in my Eth kernel by writing it all in AMD GCN assembly. AMD gives you the power to exploit the hardware to the fullest, while Nvidia does not. PTX assembly is not ISA, and it's not good enough sometimes.

Which got you all of 1-2% more then Genoil's miner? ~_~

I remember when Carmack chased the dream of getting every last ounce of performance out of Xbox 360s and PS3s with Rage. Pretty much was a waste of time when there is something better available.

[bensam1231]

again it's easy to do 240MH on fury when they consume 270w, not impressed

currently you can do 220 with a 1070 at half the wattage, if someone can check the utilizzation on lbry, if it is not 99% then there are probably some optimizations to do, otherwise not so much

i would like to see wolfo working on nivida some times, to see what he can really pull off, or if this is because you can do more with amd than nvidia...something is telling me that it is the latter...

AMD doesn't have the luxuries that Nvidia has for developers, which is one reason I liked it in the beginning. But what AMD does offer is unprecedented control, which I exploited in my Eth kernel by writing it all in AMD GCN assembly. AMD gives you the power to exploit the hardware to the fullest, while Nvidia does not. PTX assembly is not ISA, and it's not good enough sometimes.

Which got you all of 1-2% more then Genoil's miner? ~_~

I remember when Carmack chased the dream of getting every last ounce of performance out of Xbox 360s and PS3s with Rage. Pretty much was a waste of time when there is something better available.

Actually, on compute-limited GPUs, I get more. Oh, and I DROPPED power usage, don't forget that.

Besides, how's you're CUDA miner doing on Eth?

As good as it's been doing for the last six months? Kinda why I've been pushing for Dual mining, because there is a diminishing return for the amount of work people are putting into things and getting relatively nothing out of it (compared to a dual miner). Not currently mining Eth though.

[Amph]

GTX 1070 is 5.25GH/s on XVC (Blake-256 8 round), pulling 150W - this gives it a MH/s/W value of 35MH/s/W.

I'd like to stress that it's 14nm. With my full-custom design on one of my 28nm FPGAs, I get 2.1GH/s at 24W - this gives it an MH/s/W value of 87.5MH/s/W.

As it is, this fight is one-sided. If they had been manufactured on the same node, it wouldn't be a fight - it would be an execution.

well not fair to compare a gpu with fpga, fpga can do well one thing at time, then you need to reprogram it, gpu can do multiple things

it will consume more energy because of that, if gpu were specialized only on mining, they would be just asic, so yes it's not all about nm productive process

[bensam1231]

GTX 1070 is 5.25GH/s on XVC (Blake-256 8 round), pulling 150W - this gives it a MH/s/W value of 35MH/s/W.

I'd like to stress that it's 14nm. With my full-custom design on one of my 28nm FPGAs, I get 2.1GH/s at 24W - this gives it an MH/s/W value of 87.5MH/s/W.

As it is, this fight is one-sided. If they had been manufactured on the same node, it wouldn't be a fight - it would be an execution.

well not fair to compare a gpu with fpga, fpga can do well one thing at time, then you need to reprogram it, gpu can do multiple things

it will consume more energy because of that, if gpu were specialized only on mining, they would be just asic, so yes it's not all about nm productive process

Yeah, but my point was, from a mining perspective, a GPU and an FPGA can both mine many algos. The FPGA may be somewhat more restricted in selection, but it can still switch. So comparing raw hash/watt as a measure of merit is faulty, unless what I pointed out holds as well.

FPGAs are in a completely different class... You could lump ASICs into that comparison too. They also can mine multiple algos... They just have to be built from the ground up each and every time. To that extent, so do miners for GPUs (depending on how different the algo is from other ones already made), but the time requirement is quite a bit different.

[bensam1231]

GTX 1070 is 5.25GH/s on XVC (Blake-256 8 round), pulling 150W - this gives it a MH/s/W value of 35MH/s/W.

I'd like to stress that it's 14nm. With my full-custom design on one of my 28nm FPGAs, I get 2.1GH/s at 24W - this gives it an MH/s/W value of 87.5MH/s/W.

As it is, this fight is one-sided. If they had been manufactured on the same node, it wouldn't be a fight - it would be an execution.

well not fair to compare a gpu with fpga, fpga can do well one thing at time, then you need to reprogram it, gpu can do multiple things

it will consume more energy because of that, if gpu were specialized only on mining, they would be just asic, so yes it's not all about nm productive process

Yeah, but my point was, from a mining perspective, a GPU and an FPGA can both mine many algos. The FPGA may be somewhat more restricted in selection, but it can still switch. So comparing raw hash/watt as a measure of merit is faulty, unless what I pointed out holds as well.

FPGAs are in a completely different class... You could lump ASICs into that comparison too. They also can mine multiple algos... They just have to be built from the ground up each and every time. To that extent, so do miners for GPUs (depending on how different the algo is from other ones already made), but the time requirement is quite a bit different.

ASICs can't mine multiple algos unless they're made to from the start. You can buy an FPGA ONCE and reprogram it - a GPU is closer to this. You don't have to get a new GPU every algo.

Sure, but often times you have to completely reprogram the thing from the ground up. They're both in a different class of products from GPUs.

[bensam1231]

GTX 1070 is 5.25GH/s on XVC (Blake-256 8 round), pulling 150W - this gives it a MH/s/W value of 35MH/s/W.

I'd like to stress that it's 14nm. With my full-custom design on one of my 28nm FPGAs, I get 2.1GH/s at 24W - this gives it an MH/s/W value of 87.5MH/s/W.

As it is, this fight is one-sided. If they had been manufactured on the same node, it wouldn't be a fight - it would be an execution.

well not fair to compare a gpu with fpga, fpga can do well one thing at time, then you need to reprogram it, gpu can do multiple things

it will consume more energy because of that, if gpu were specialized only on mining, they would be just asic, so yes it's not all about nm productive process

Yeah, but my point was, from a mining perspective, a GPU and an FPGA can both mine many algos. The FPGA may be somewhat more restricted in selection, but it can still switch. So comparing raw hash/watt as a measure of merit is faulty, unless what I pointed out holds as well.

FPGAs are in a completely different class... You could lump ASICs into that comparison too. They also can mine multiple algos... They just have to be built from the ground up each and every time. To that extent, so do miners for GPUs (depending on how different the algo is from other ones already made), but the time requirement is quite a bit different.

ASICs can't mine multiple algos unless they're made to from the start. You can buy an FPGA ONCE and reprogram it - a GPU is closer to this. You don't have to get a new GPU every algo.

Sure, but often times you have to completely reprogram the thing from the ground up. They're both in a different class of products from GPUs.

You do realize GPUs are pretty much the same, except they expose an instruction set, correct?

Something about memory, horsepower too (computational units), instruction sets they support, and operating environment. Even if you can do one thing really well with a FPGA (much like a ASIC), that doesn't mean it'll do everything else pretty much equally as ewll. There is a reason FPGAs have always been the stepping stone to ASICs. Because if you're going to take enough time to program for a FPGA, you can just take that one step further and start designing the chip too, which adds a lot more flexibility when it comes to efficiency and raw horsepower (more of whatever you need to produce a certain amount of hashrate, less of whatever isn't being used) and allows your clientele to easily implement them (a box you plug in). The level of expertise you need for each of those goes up quite a bit hoping from GPU > FPGA > ASIC.

Like I said, there is pretty much three classes, GPUs, ASICs, and FPGAs. FPGAs are like the experimentation ground for ASICs. There are CPUs too, but GPUs can do almost everything CPUs can better, especially when it comes to cryptos.

[fenomenhaa]

My results on lbry.

750 ti with o/c :49-50 mh/s
970 with o/c  : 131-132mh/s
1070  with o/c (2000mhz core):222-224mh/s  

7970-280x o/c : 50-51mh/s
290 with oc     :66 -67 mh/s

[Velgelm]

Nvidia seems good!
What miner are you using for Nvidia cards?

[Amph]

My results on lbry.

750 ti with o/c :49-50 mh/s
970 with o/c  : 131-132mh/s
1070  with o/c (2000mhz core):222-224mh/s  

7970-280x o/c : 50-51mh/s
290 with oc     :66 -67 mh/s

what is your usage on the 1070? 99%?

Nvidia seems good!
What miner are you using for Nvidia cards?

there is no open source for nvidia yet, so i guess he is using the private one

[Epsylon3]

"preview" one... or "test" one if you prefer

else yiimp has internet issues since a few hours... sorry about that

[fenomenhaa]

My results on lbry.

750 ti with o/c :49-50 mh/s
970 with o/c  : 131-132mh/s
1070  with o/c (2000mhz core):222-224mh/s  

7970-280x o/c : 50-51mh/s
290 with oc     :66 -67 mh/s

what is your usage on the 1070? 99%?

Nvidia seems good!
What miner are you using for Nvidia cards?

there is no open source for nvidia yet, so i guess he is using the private one

1070 @ %94-95 it seems possible to improvee.Sure private miner.

[pallas]

My results on lbry.

750 ti with o/c :49-50 mh/s
970 with o/c  : 131-132mh/s
1070  with o/c (2000mhz core):222-224mh/s  

7970-280x o/c : 50-51mh/s
290 with oc     :66 -67 mh/s

what is your usage on the 1070? 99%?

Nvidia seems good!
What miner are you using for Nvidia cards?

there is no open source for nvidia yet, so i guess he is using the private one

1070 @ %94-95 it seems possible to improvee.Sure private miner.

Tpruvot will publish it in two days, then I will have a look and see if I can improve it further, and eventually push the commits.
Recently he's done a very good job so don't hold your breath ;-)