[ANN]pledge for the release of 100% faster cuda 7.5 nvidia miner for neoscrypt

[djm34]

As far as I understand (please correct me, djm34, if I'm wrong), the 50% gain is based on the latest opensource kernel by djm34, NOT the one on the sp_ fork, which includes about 10% improvement by sp_ and another 10% by myself.
So the gain at the end should be about 30%.

don't know for sure actually, if you can tell me what is the hashrate of the opensource code (or which sp version to use, I can do a comparison, I think sp change on blake mixing was incorporated before I gave a quote)

[Al_may]

As far as I understand (please correct me, djm34, if I'm wrong), the 50% gain is based on the latest opensource kernel by djm34, NOT the one on the sp_ fork, which includes about 10% improvement by sp_ and another 10% by myself.
So the gain at the end should be about 30%.

don't know for sure actually, if you can tell me what is the hashrate of the opensource code (or which sp version to use, I can do a comparison, I think sp change on blake mixing was incorporated before I gave a quote)

with SP_ ver .74 my 970's get 563 with his ver 7.7 I get 368 Kh/s on neoscrypt

[Grout]

87a38017a306d45152261735da5a2c21473bd7891886783cd960b05bb2f45a67
Even if you don't reach the goal, consider this a thank you for your previous work and an incentive to opensource a little more, and troll a little less ;-)

[djm34]

As far as I understand (please correct me, djm34, if I'm wrong), the 50% gain is based on the latest opensource kernel by djm34, NOT the one on the sp_ fork, which includes about 10% improvement by sp_ and another 10% by myself.
So the gain at the end should be about 30%.

don't know for sure actually, if you can tell me what is the hashrate of the opensource code (or which sp version to use, I can do a comparison, I think sp change on blake mixing was incorporated before I gave a quote)

with SP_ ver .74 my 970's get 563 with his ver 7.7 I get 368 Kh/s on neoscrypt

ok so it is even more than 50% (I don't have a 970, I only have 980's)

[jakals]

As far as I understand (please correct me, djm34, if I'm wrong), the 50% gain is based on the latest opensource kernel by djm34, NOT the one on the sp_ fork, which includes about 10% improvement by sp_ and another 10% by myself.
So the gain at the end should be about 30%.

don't know for sure actually, if you can tell me what is the hashrate of the opensource code (or which sp version to use, I can do a comparison, I think sp change on blake mixing was incorporated before I gave a quote)

with SP_ ver .74 my 970's get 563 with his ver 7.7 I get 368 Kh/s on neoscrypt

ok so it is even more than 50% (I don't have a 970, I only have 980's)

will you look at adding support for cuda 2.1 djm ? or is it not applicable

[djm34]

87a38017a306d45152261735da5a2c21473bd7891886783cd960b05bb2f45a67
Even if you don't reach the goal, consider this a thank you for your previous work and an incentive to opensource a little more, and troll a little less ;-)

Thanks for your support, can't promise much on the trolling part though   (which is usually a reaction on something I consider as trolling)
Djm

[pallas]

As far as I understand (please correct me, djm34, if I'm wrong), the 50% gain is based on the latest opensource kernel by djm34, NOT the one on the sp_ fork, which includes about 10% improvement by sp_ and another 10% by myself.
So the gain at the end should be about 30%.

don't know for sure actually, if you can tell me what is the hashrate of the opensource code (or which sp version to use, I can do a comparison, I think sp change on blake mixing was incorporated before I gave a quote)

with SP_ ver .74 my 970's get 563 with his ver 7.7 I get 368 Kh/s on neoscrypt

ok so it is even more than 50% (I don't have a 970, I only have 980's)

Hmmmm no.... your version should do 650-700 while mine does 560 on the 970.

[djm34]

As far as I understand (please correct me, djm34, if I'm wrong), the 50% gain is based on the latest opensource kernel by djm34, NOT the one on the sp_ fork, which includes about 10% improvement by sp_ and another 10% by myself.
So the gain at the end should be about 30%.

don't know for sure actually, if you can tell me what is the hashrate of the opensource code (or which sp version to use, I can do a comparison, I think sp change on blake mixing was incorporated before I gave a quote)

with SP_ ver .74 my 970's get 563 with his ver 7.7 I get 368 Kh/s on neoscrypt

ok so it is even more than 50% (I don't have a 970, I only have 980's)

will you look at adding support for cuda 2.1 djm ? or is it not applicable

I can look, however I can't promise; I tried on the previous version and didn't succeed.

[djm34]

No. After Djm34 has released his new sourcecode I will add a few percent and add it to my fork as I usually do. Then you can donate.

thank you for allowing me to test the moderation 

[ldw-com]

Only 88% more in donations needed. Almost there! 

[djm34]

Only 88% more in donations needed. Almost there!  

tbh, considering the pledge only started yesterday, this is rather good (and someone donated 0.1btc before the start of the pledge)

[ldw-com]

Only 88% more in donations needed. Almost there!  

tbh, considering the pledge only started yesterday, this is rather good (and someone donated 0.1btc before the start of the pledge)

Ah okay! That's better! Well, if you keep this thread alive, i'm sure you'll make it Good luck!

[djm34]

Only 88% more in donations needed. Almost there! 

tbh, considering the pledge only started yesterday, this is rather good (and someone donated 0.1btc before the start of the pledge)

Ah okay! That's better! Well, if you keep this thread alive, i'm sure you'll make it Good luck!

Thanks 
djm

[bathrobehero]

Any chance we get numbers on a 970? Also, are these stock figures or with OC?

[djm34]

Any chance we get numbers on a 970? Also, are these stock figures or with OC?

Those numbers are with OC actually, overclocking the memory clock helps a bit like with all memory hard algo.
Regarding the 970, I don't have one... so it is difficult to say...

[chrysophylax]

Any chance we get numbers on a 970? Also, are these stock figures or with OC?

Those numbers are with OC actually, overclocking the memory clock helps a bit like with all memory hard algo.
Regarding the 970, I don't have one... so it is difficult to say...

djm34 ...

is there a way we can gauge how far into the pledge we are as time goes on? ...

maybe an update in the op once every few days or something? ...

#crysx

[djm34]

Any chance we get numbers on a 970? Also, are these stock figures or with OC?

Those numbers are with OC actually, overclocking the memory clock helps a bit like with all memory hard algo.
Regarding the 970, I don't have one... so it is difficult to say...

djm34 ...

is there a way we can gauge how far into the pledge we are as time goes on? ...

maybe an update in the op once every few days or something? ...

#crysx

sure, just need to use the block explorer: https://blockchain.info/address/16UoC4DmTz2pvhFvcfTQrzkPTrXkWijzXw (+0.1btc actually)
I can add that link to the OP

[djm34]

As far as I understand (please correct me, djm34, if I'm wrong), the 50% gain is based on the latest opensource kernel by djm34, NOT the one on the sp_ fork, which includes about 10% improvement by sp_ and another 10% by myself.
So the gain at the end should be about 30%.

don't know for sure actually, if you can tell me what is the hashrate of the opensource code (or which sp version to use, I can do a comparison, I think sp change on blake mixing was incorporated before I gave a quote)

with SP_ ver .74 my 970's get 563 with his ver 7.7 I get 368 Kh/s on neoscrypt

ok so it is even more than 50% (I don't have a 970, I only have 980's)

Hmmmm no.... your version should do 650-700 while mine does 560 on the 970.

ok I did a few check with the sp/pallas version (74), this new version is "only" 25% faster at this point.

[pallas]

Djm34, does this version work fast on cuda 7.5? How does it compare to itself compiled on 6.5? Thanks.

[Grout]

87a38017a306d45152261735da5a2c21473bd7891886783cd960b05bb2f45a67
Even if you don't reach the goal, consider this a thank you for your previous work and an incentive to opensource a little more, and troll a little less ;-)

Thanks for your support, can't promise much on the trolling part though   (which is usually a reaction on something I consider as trolling)

Dammit! Refund please! :-D